Development of Enzyme-Linked Immunosorbent Assays for the Detection of Mutagenic Metabolites of the Herbicide Alachlor

University of Massachusetts Amherst ScholarWorks@UMass Amherst

Doctoral Dissertations 1896 - February 2014

1-1-1998

Development of enzyme-linked immunosorbent assays for the detection of mutagenic metabolites of the herbicide alachlor.

Daniel M. Tessier University of Massachusetts Amherst

Follow this and additional works at: https://scholarworks.umass.edu/dissertations_1

Recommended Citation Tessier, Daniel M., "Development of enzyme-linked immunosorbent assays for the detection of mutagenic metabolites of the herbicide alachlor." (1998). Doctoral Dissertations 1896 - February 2014. 5666. https://scholarworks.umass.edu/dissertations_1/5666

This Open Access Dissertation is brought to you for free and open access by ScholarWorks@UMass Amherst. It has been accepted for inclusion in Doctoral Dissertations 1896 - February 2014 by an authorized administrator of ScholarWorks@UMass Amherst. For more information, please contact [email protected].

DEVELOPMENT OF ENZYME-LINKED IMMUNOSORBENT ASSAYS FOR THE DETECTION OF MUTAGENIC METABOLITES OF THE HERBICIDE ALACHLOR

A Dissertation Presented

DANIEL M. TESSIER

Submitted to the Graduate School of the University of Massachusetts Amherst in partial fullfillment of the requirements for the degree of

DOCTOR OF PHILOSOPHY

February 1998

Department of Entomology © Copyright Daniel M. Tessier 1998

A Dissertation Presented

DANIEL M. TESSIER

Approved as to style and content by: yC A. oLri

Peter C. Uden, Member

T. Michaej/Peters, Department Head Department of Entomology

ACKNOWLEDGEMENTS

I thank Professor John M. Clark for directing my dissertation research and for many years of invaluable support and guidance. Professor Nordin, Professor Uden and

Professor Yin are acknowledged for their input and willingness to serve on my dissertation committee. Dr. Bruce Hammock of UC Davis provided critical insight into the hapten conjugation strategy. I thank Dr. E. Michael Thurman of the US Geological

Survey for providing groundwater samples for the validation aspects of this research.

To the gang at the Pesticide Lab; Ray, Andy, Gerry, Jeff, thanks so much for creating a collegial, highly satisfying (if aesthetically decrepit) place to work. Carry on, men. To the Toxicology group; Steve, Jessica, Aiguo, Yoon, Scott, Kosea and the rest of the new

Tox Generation, consider the torch passed. Thanks for everything.

v ABSTRACT

DEVELOPMENT OF ENZYME-LINKED IMMUNOSORBENT ASSAYS FOR THE DETECTION OF MUTAGENIC METABOLITES OF THE HERBICIDE ALACHLOR

FEBRUARY 1998

DANIEL M. TESSIER

B.S., UNIVERSITY OF MASSACHUSETTS AMHERST

M.S., UNIVERSITY OF MASSACHUSETTS AMHERST

Ph.D., UNIVERSITY OF MASSACHUSETTS AMHERST

Directed by: Professor J. Marshall Clark

The herbicide alachlor is one of the most widely used pesticides in the world; over 52

million pounds are applied to U.S. croplands annually. The acetanilide compounds 2-

chloro-2',6'-diethylacetanilide (CDA) and 2-hydroxy-2',6'-diethylacetanilide (HDA) are

environmental degradative products of alachlor. CDA, HDA and alachlor are ground and

surface water contaminants; CDA and HDA are mutagenic in the Salmonella / microsome

assay. There is a paucity of data on the environmental fate of CDA and HDA.

The development of two competitive enzyme-linked immunosorbent assays (cELISA)

for the detection of CDA and HDA is reported. cELISA3 is specific for CDA with a

detection range of 0.015 to 10 pg/ml. Solid phase extraction of CDA residues from

aqueous samples gives a 1000-fold concentration factor resulting in an effective detection

limit of 15 pg/ml. cELISA4 is specific for both CDA and HDA in combination, with a detection range of 0.01 to 10 pg/ml. Solid phase extraction of aqueous samples prior to cELISA analysis results in an effective detection limit of 10 pg/ml. Chloroacetanilide

vi herbicides and other alachlor metabolites that may be present in environmental samples do not interfere with the detection of CDA and HDA.

cELISA3, cELISA4 and the antisera they are based on provide a means of studying the environmental fate of CDA and HDA through a variety of analytical strategies. TABLE OF CONTENTS

Page ACKNOWLEDGEMENTS. v

ABSTRACT. vi

LIST OF TABLES. xii

LIST OF FIGURES. xiii

LIST OF ABBREVIATIONS. xvi

Chapter

I. INTRODUCTION. 1

A. The Herbicide Alachlor. 1

1. Status of Alachlor Use. 1 2. Magnitude and Significance of Environmental Contamination. 1

B. Environmental Metabolism of Alachlor. 3 C. Mutagenicity of Alachlor Metabolites. 7 D. Cross-Reactivity of CDA and HDA Mutagens in Alachlor Immunoassays. 11 E. Utility of an Alachlor Metabolite Immunoassay. 12

II. SYNTHESIS OF HAPTENIC DERIVATIVES FOR THE PREPARATION OF CDA AND HDA IMMUNOGENS. 14

A. Chemicals. 14 B. Apparatus. 14 C. Synthesis of Haptens. 15

1. Synthesis of Hapten C2-CD A. 15

a. 2,6-Diethylaniline-V-ethyl acetate (2). 15 b. 2,6-Diethylaniline-V-acetic acid (3). 18 c. 2-Chloro-2’ ,6’-diethyl(V-acetic acid)acetanilide (4). 18

vm 19 2. Synthesis of Hapten C4-CDA

a. 2,6-Diethylaniline-A-ethylbutyrate (5). 1 * b. 2,6-Diethylaniline-iV-butyric acid (6). 22 c. 2-Chloro-2\6’-diethyl(A-butyric acid)acetanilide (7). 22

3. Synthesis of Hapten Phe-CDA. 22

a. 3,5-Diethylphenol (9). 22 b. 3,5-Diethyl-4-nitrosophenol (10). 22 c. 2,6-Diethyl-4-hydroxyaniline (11). 22 d. 2,6-Diethyl-4-[4’-(ethoxycarbonyl)butoxy]aniline (12). 28 e. 2-Chloro-2’,6’-diethyl-4’-[4”-(ethoxycarbonyl) butoxy]acetanilide (13). 2^ f. 2-Chloro-2 ’ ,6’ -diethyl-4 ’ - [4 ” -(carboxylic acid) butoxy]acetanilide (14). 29

D. Preparation of Hapten-Protein Conjugates. 29

1. Preparation of Active Esters of the CD A Haptens. 29

' a. General method for the formation of active esters. 32 b. Structural confirmation of the active ester of C2-CDA. 32 c. Structural confirmation of the active ester of C4-CDA. 32 d. Structural confirmation of the active ester of Phe-CDA. 33

2. Conjugation of CDA Haptens to Carrier Proteins. 33

a. Conjugation of active esters. 33 b. Conjugation via N-acetyl homocysteinethiolactone. 33

III. DEVELOPMENT OF A COMPETITION ENZYME-LINKED IMMUNOSORBENT ASSAY (cELISA) FOR THE DETECTION OF CDA AND HD A. 39

A. Antibody Production. 39 B. Procedures for Enzyme Immunoassays. 39 C. Competition Enzyme-Linked Immunosorbent Assays (cELISA). 44

1. cELISA 1: Anti-BSA-C2-CDA Antisera. 44

a. Checkerboard assays. 44 b. Competition assays. 44 c. Cross-reactivity to related compounds. 45

IX 2. cELISA 2: Anti-BSA-C4-CDA Antisera 48

a. Checkerboard assays. 48 b. Competition assays. 48 c. Cross-reactivity to related compounds. 48

3. cELISA 3: Anti-BSA-Phe-CDA Antisera. 54

a. Checkerboard assays. 54 b. Competition assays. 54 c. Cross-reactivity to related compounds. 54 d. Assay optimization. 58

i. pH effects. 58 ii. Solvent effects. 61

4. cELISA 4: Anti-BSA-AHT-CDA Antisera. 62

a. Checkerboard assays. 62 b. Competition assays. 62 c. Cross-reactivity to related compounds. 63 d. Assay optimization. 66

i. pH effects. 66 ii. Solvent effects. 66

D. cELISA Validation... 70

1. Solid Phase Extraction of Aqueous Samples. 70 2. Chromatographic Analysis of Aqueous Sample Extracts. 76 3. Correlation of cELISA and Chromatographic Analysis of Aqueous Samples. 75 4. cELISA Analysis of Alachlor-Contaminated Groundwater Samples for CD A. 75

IV. DISCUSSION AND CONCLUSIONS. 82

A. Hapten Design. g2 B. The CDA and CDA/HDA cELISAs.87 C. Conclusions. gq

x APPENDICES 91

A. MASS SPECTRA. 92 B. INFRARED SPECTRA. 115 C. 1H NUCLEAR MAGENTIC RESONANCE SPECTRA. 140

BIBLIOGRAPHY. 145

XI LIST OF TABLES

Table Pa8e

1. Selected EPA Leaching Criteria and Chemical Parameters for Alachlor.2

2. Hapten Densities of Immunogens and Plate-Coating Antigens. 34

3. Hapten-Carrier Conjugates Used for Immunization and Plate Coating.40

4. Antisera Designations. 40

5. Checkerboard Assay for Antisera RAX1 versus BSA-C2-CDA and HSA-C2-CDA Plate-Coating Antigens. 45

6. Checkerboard Assay for the Detection of BSA-C4-CDA by Antisera RAX5, RAX6 and RAX7. 49

7. Cross-Reactivity (Reported as IC50) of Chloroacetanilide Herbicides to Antiserum RAX7. 49

8. Checkerboard Assay for the Detection of OVA-Phe-CDA by Antisera , RAX10 and RAX11. 55

9. Cross-Reactivity of RAX11 to Chloracetanilide Herbicides and Related Compounds. 58

10. cELISA3 Signal Attenuation Due to the Effect of Solvents on Antibody Binding. 61

11. Checkerboard Assay for the Detection of OVA-AHT-CDA by Antiserum RAX8. 63

12. Cross-Reactivity of Antiserum RAX8 to Chloracetanilide Herbicides and Related Compounds. 66

13. cELISA4 Signal Attenuation Due to the Effect of Solvents on Antibody Binding. 69

14. Recovery of CD A and HD A from Fortified Water Samples by Solid Phase Extraction... 71

15. Analysis of Alachlor-Contaminated Groundwater Samples 77 LIST OF FIGURES

Figure Page

1. Environmental degradative products of alachlor. 6

2. Proposed mechanism of action of CD A and HD A as toxic electrophiles 10

3. Synthesis of hapten C2-CDA (4) containing a 2-carbon spacer group attached to the nitrogen moiety. 17

4. Synthesis of hapten C4-CDA (7) containing a 4-carbon spacer group attached to the nitrogen moiety. 21

5. Synthesis of 2,6-diethyl-4-hydroxyaniline. 24

6. Synthesis of hapten Phe-CDA (14) containing a 5-carbon spacer group attached to the aromatic ring. 26

7. Formation of active esters of C2-CDA, C4-CDA and Phe-CDA via N-hydroxysuccinimide... 31

8. Direct coupling of CDA to protein carrier molecules via N-acetyl homocysteine thiolactone (AHT). 36

9. The hapten-carrier conjugates used to immunize rabbits for the production of CDA and CDA / HDA specific antibodies. 38

10. Antibody capture format used for all competition enzyme-linked immuno¬ sorbent assays (cELISA). 42

11. Competitive inhibition of antiserum RAX1 by CDA and alachlor.47

12. Competitive inhibition of antiserum RAX7 by CDA and alachlor.51

Structures of chloroacetanilide herbicides and related compounds tested for cross-reactivity. 53

14. Competitive inhibition of antiserum RAX11 by CDA. 57

15. Effect of aqueous sample pH on binding of antiserum RAX11.60

16. Competitive inhibition of antiserum RAX8 by CDA and HDA.65

17. Effect of aqueous sample pH on binding of antiserum RAX8.68

xm 18. Correlation of fortified sample CD A concentration with CD A concentration detected by cELISA3. 73

19. Correlation of fortified sample CD A and HD A concentration with CDA and HDA concentration detected by cELISA4. 73

20. Correlation of CDA detection by cELISA3 with GC/MS. 7 9

21. Correlation of CDA and HDA detection by cELISA4 with GC/MS. 81

22. Candidate attachment sites for covalent linkage of CDA to carrier protein molecules. 8 3

A1. Full-scan mass spectrum (50-550 mass units) of 2-chloro-2’,6’-diethylacetanilide (CDA). 94

A2. Full-scan mass spectrum (50-550 mass units) of 2-hydroxy-2’,6’-diethylacetanilide (HDA)... 96

A3. Full-scan mass spectrum (50-550 mass units) of 2,6-diethylaniline- N-ethyl acetate (2). 98

A4. Full-scan mass spectrum (50-550 mass units) of 2,6-diethylaniline-N- acetic acid (3). 100

A5. Full-scan mass spectrum (50-550 mass units) of 2-chloro-2’,6’-diethyl (N-acetic acid)acetanilide (4). 102

A6. Full-scan mass spectrum (50-550 mass units) of 2,6-diethylaniline-N- ethyl-butyrate (5). 104

A7. Full-scan mass spectrum (50-550 mass units) of 3,5-Diethylphenol (9). 106

A8. Full-scan mass spectrum (50-550 mass units) of 2,6-diethyl-4- hydroxyaniline (11). 108

A9. Full-scan mass spectrum (50-550 mass units) of 2,6-diethyl-4-[4’- (ethoxy-carbonyl)butoxy] aniline (12). 110

A10. Full-scan mass spectrum (50-550 mass units) of 2-chloro-2’,6’- diethy 1-4’-[4”-(ethoxycarbonyl)butoxy] acetanilide (13). 112

xiv All. Full-scan mass spectrum (50-550 mass units) of the methyl ester of 2- chloro- 2 ’ ,6 ’ -diethyl-4’ - [4”(carboxylie acid)butoxy] acetanilide (14).

Bl. IR spectrum (neat) of 2,6-diethylaniline-N-ethylacetate (2). 117

B2. IR spectrum (neat) of 2,6-diethylaniline-N-acetic acid (3). 119

B3. IR spectrum (neat) of 2-chloro-2’,6’-diethyl(N-acetic acid)acetanilide (4).. 121

B4. IR spectrum (neat) of 2,6-diethylaniline-N-ethyl-butyrate (5). 123

B5. IR spectrum (KBr pellet) of 3,5-Diethylphenol (9). 125

B6. IR spectrum (KBr pellet) of 3,5-Diethyl-4-nitrosophenol. 127

B7. IR spectrum (neat) of 2,6-diethyl-4-[4’-(ethoxy-carbonyl)butoxy] aniline (12). 129

B8. IR spectrum (Nujol) of 2-chloro-2’,6’-diethyl-4’-[4’’-(ethoxycarbonyl) butoxy] acetanilide (13). 131

B9. IR spectrum (KBr pellet) of 2-chloro-2’,6’-diethyl-4’-[4’’(carboxylic acid) butoxy] acetanilide (14). 133

BIO. IR spectrum (neat) of the active ester of C2-CD A. 135

B11. IR spectrum (neat) of the active ester of C4-CDA. 137

B12. IR spectrum (neat) of the active ester of Phe-CDA. 139

Cl. ]H NMR spectrum (CDCI3, 200 Mhz) of 2,6-diethylaniline-N-ethylacetate (2). 142

C2. !H NMR spectrum (CDCI3, 200 Mhz) of the active ester of C2-CDA. 144

xv LIST OF ABBREVIATIONS

AHT N-acetyl homocystein thiolactone

BSA bovine serum albumin

dH20 distilled water

DMSO dimethylsulfoxide

GC/MS gas chromatography / mass selective detector

]H NMR proton nuclear magnetic resonance spectroscopy

HRP horseradish peroxidase

HSA human serum albumin

IR infrared spectroscopy

OD optical density

OVA ovalbumin (chicken egg white albumin)

PBS phosphate-buffered saline

PBST phosphate-buffered saline amended with 0.05% Tween-20

THF tetrahydrofuran

xvi CHAPTER I

INTRODUCTION

A. The Herbicide Alachlor

1. Status of Alachlor Use

The chloroacetanilide herbicide alachlor is one of the most extensively-used agrochemicals in the United States of America (USA). It is a selective, preemergent herbicide used on com, soybean, sorghum and other crops for broadleaf weed control.

Alachlor is manufactured by Monsanto Agricultural Company (St. Louis, MO) under the trade names Lasso™, Lasso II™ and Lasso Micro-Tech™ and is also formulated in mixtures with atrazine, glyphosate, propanil, and other herbicides. Approximately 95 million pounds were applied annually to American croplands during the peak of alachlor use in the late 1970s to early 1980s (EPA 1987, Sun 1986). Current use patterns have decreased to approximately 52 million pounds applied annually (Gianessi and Anderson

1995). However, alachlor still ranks third, after atrazine (72 million pounds applied / year) and metolachlor (59 million pounds applied / year), in total usage per year for all pesticides. In the USA, alachlor is applied to 30% of sweetcom acreage, 20% of feed com acreage, 10% of soybean and sorghum acreage, and less than 10% of cotton and peanut acreage (Gianessi and Anderson 1995).

2. Magnitude and Significance of Environmental Contamination

Numerous studies, conducted by the U.S. Environmental Protection Agency (EPA),

U.S. Department of Agriculture (USDA), U.S. Geological Survey (USGS), and Monsanto

1 have shown alachlor to be a commonly detected contaminant in ground and surface

waters. It is moderately soluble in water and exhibits other chemical properties that

contribute to its mobility in the environment (Table 1). Contamination of ground and

surface water by alachlor occurs via leaching through the soil profile (groundwater) and

overland flow (surface water) in the dissolved state. Contaminated groundwater also

significantly contributes to surface water contamination (Pereira and Rostad 1990; Pereira

et al. 1992; Pereira and Hostettler 1993; Thurman et al. 1996). Volatilization from the

soil surface results in detectable residues of alachlor in fog and rainwater, where they are

dispersed and redeposited on land and water surfaces (Goolsby et al. 1997).

Table 1. Selected EPA Leaching Criteria and Chemical Parameters for Alachlor

Parameter EPA Leacher Alachlor

Water solubility >30 mg/L 242 mg/L

Kd1 <5 0.3 -3.7

Hydrolysis half-life > 180 d stable

Photolysis half-life >3 d stable

Soil half-life > 14-21 d 6-23 d 1. Soil/water adsorption coefficient. Refs: Montgomery 1993; Wienhold and Gish 1994.

Groundwater contamination, typically in the low part per billion range, has been

reported in at least 24 of 34 states where alachlor is used (Jacoby et al. 1992). The

frequency of detection is 0.1 to 5% of all wells tested. In studies that focused on areas of heavy alachlor use, however, upwards of 75% of private wells are contaminated with alachlor (Chesters et al. 1989). Surface waters are also significantly impacted by alachlor.

The Mississippi River, the major water drainage system in the central USA, receives a

2 complex mixture of agrochemicals from this primarily agricultural region. It is estimated

that the Mississippi River annually transports 18 tons of alachlor into the Gulf of Mexico

(Pereira and Hostettler 1993).

Although the acute mammalian toxicity of alachlor is relatively low (LD50 = 930 mg/kg,

oral, rat), it is a proven animal carcinogen. The EPA designates alachlor as a class B2

carcinogen or "probable human" carcinogen. Therefore, the contamination of ground and

surface waters by alachlor may be regarded as toxicologically significant from the

perspective of chronic human exposure via drinking water supplies.

The presence of alachlor in ground and surface water, coupled with its oncogenicity,

resulted in an outright banning of this herbicide by Canada in 1985 (Hoberg 1990).

Although registered as a ‘"restricted use” agrochemical in the USA, alachlor is still one of

the most heavily applied pesticides and a significant ground and surface water

contaminant.

B. Environmental Metabolism of Alachlor

The study of the chemistry, fate, transport and analysis of environmental metabolites of pesticides has historically not received as much attention as research on the parent pesticides themselves. This has been due to the large number of potential analytes arising from the breakdown of any given parent compound, coupled with a backlog of environmental research still needed for existing and newly developed pesticides.

Degradation in the environment, whether biotic, abiotic or photolytic, most commonly results in compounds that retain the structural core of the parent pesticide. These

3 degradation products may retain some of the activity of the parent and / or exhibit mechanisms of toxicity unrelated to the mode of action of the parent (Barrett 1996).

Metabolites are typically more stable and more polar than the parent compound and therefore more mobile in the environment. In this manner, metabolites can have an even greater impact on ground and surface waters than the parent compound. Except for the sulfoxide and sulfone metabolites of aldicarb, there are no established tolerances for pesticide metabolites in water (Barbash and Resek 1996).

The biotic, primarily microbial, degradation of alachlor results in a variety of aromatic amine metabolites that also have been detected in ground and surface waters (Fig. 1;

Potter and Carpenter 1995, Pereira and Hostettler 1993, Pereira et al. 1992, Pereira and

Rostad 1990, Aizawa 1982, Tiedje and Hagedorn 1975). Abiotic oxidative metabolism plays only a minimal role in degradation and mineralization does not occur (Novick

1986). Thus, movement of these metabolites out of the microbially active zone of the soil profile (i.e., the rhyzosphere) may result in some of these metabolites existing as long¬

term or terminal residues with the potential to contaminate ground and surface waters.

Alachlor has been detected in the rhyzosphere for up to one year after application

(Heyer and Stan 1995), thus providing a long-term input for metabolite formation. Potter

and Carpenter (1995) identified 20 distinct compounds in groundwater samples that were

postulated to arise from alachlor degradation. The major metabolites of alachlor that

have been studied to date are alachlor-ethanesulfonic acid [ESA; (Thurman et al. 1996,

Baker et al. 1993)], 2-chloro-2',6'-diethylacetanilide [CDA; (Potter and Carpenter 1995,

Pereira and Hostettler 1993, Pereira et al. 1992, Pereira and Rostad 1990)] and

4 Figure 1. Environmental degradative products of alachlor. I: alachlor; II: 2-hydroxy-2',6'- diethyl-N-methoxymethylacetanilide; III: 2,6-diethyl-N-methoxymethylacetanilide; IV:

2,6-diethyl-N-methoxymethylaniline; V: 2-chloro-2',6'-diethylacetanilide; VI: 2-hydroxy-

2',6'-diethylacetanilide; VII: 2,6-diethylacetanilide; VIII: 2,6-diethylaniline.

5 o II ch3och2^ xch2ci

I II 2-hydroxy-2',6'-diethylacetanilide [HDA; (Potter and Carpenter 1995)]. These compounds exhibit environmental mobility and have been detected in both ground and surface waters at part per trillion levels. Nevertheless, the rate of formation, persistence, movement and ultimate environmental fate for these and other metabolites of alachlor is unknown.

It is also unknown how evolving agricultural practices are influencing metabolite formation. For example, controlled-release formulations are used to reduce herbicide leaching (Lee and Weber 1993). Controlled-release formulations enhance the retention of pesticides in the rhyzosphere, providing enhanced efficacy and allowing a reduction in application rates. The rhyzosphere is the primary site of pesticide degradation, however, so this strategy may result in elevated levels of metabolite formation. Another example is addition of conditioners or adjuvants to irrigation water for the reduction of soil erosion

(Sojka and Lentz 1993; Weitersen et al. 1993). This practice may also enhance the retention of pesticides in the rhyzosphere by modifying the sorptive characteristics of the soil, resulting in elevated levels of metabolite formation. In other reports, soil microbes have been shown to selectively use adjuvants as carbon sources leaving the pesticide unchanged and available for groundwater contamination. Thus, the overall influence of soil conditioning practices and controlled-release formulations must be assessed to insure that the net result is not an increase in metabolites reaching ground and surface water sources, particularly when they are toxic themselves.

C. Mutagenicity of Alachlor Metabolites

The acute mammalian toxicities of herbicides and fungicides are relatively low

7 compared to insecticides. However, there is little information to describe the toxic potential of their metabolites. Chronic human exposure can occur through contaminated ground or surface waters that serve as potable water supplies. Soil-dwelling and aquatic organisms can also be chronically exposed to these metabolites, resulting in the possibility of bioaccumilation and biomagnification.

The chronic toxicity potential of environmental metabolites of alachlor was evaluated by Tessier and Clark (1995) utilizing the Salmonella / microsome assay (Maron and

Ames 1983) and the micronucleus assay (Hayashi et al. 1990). Two metabolites, 2- chloro-2',6'-diethylacetanilide (CDA, V, Fig. 1) and 2-hydroxy-2',6'-diethylacetanilide

(HDA, VI, Fig. 1), were determined to be weakly mutagenic to Salmonella strain TA100.

This strain is sensitive to point mutations. Mutagenicity of CDA was independent of, but enhanced by, microsomal monooxygenase bioactivation, whereas HDA was dependent on bioactivation to elicit its mutagenic effects. Thus, CDA is a direct-acting mutagen and

HDA is a promutagen.

It was hypothesized that the mechanism of action of CDA and HDA is through bioactivation to electrophilic species, which then covalently react with nucleophilic

DNA/RNA to form alkylated or otherwise structurally altered forms of the genetic material (Fig 2). The glutathione S-tranferases (GST) are a family of detoxifying enzymes. They are primarily responsible for the deactivation of electrophilic species by conjugating them to glutathione and forming water-soluble, excretable adducts (Jakoby

1978). Experiments with GST showed a decrease in the mutagenicity of HDA, presumably by GST conjugation (Tessier and Clark 1995). This finding supports the

8 Figure 2. Proposed mechanism of action of CDA and HDA as toxic electrophiles. In the figure X = -Cl (in CDA) or -OH (in HDA). Cytochrome P450-based dechlorination or dehydroxylation results in electrophilic species that react with nucleophilic DNA or RNA to form point mutations.

9 x-

10 proposed mutagenic mechanism of CDA and HDA as toxic electrophiles. Since CDA caused mutations without bioactivation at high doses, other mechanisms may also play a role. Results of murine micronucleus tests indicate that these compounds do not have

clastogenic activity (Tessier and Clark 1995). Their genotoxicity is thus limited to

specific molecular interactions with DNA/RNA, forming point mutations. Since

mutagenesis commonly precedes carcinogenesis, the mutagenic metabolites of alachlor

are very likely to play a role in the carcinogenicity of alachlor. Therefore, the

quantitation of the overall environmental impact of alachlor use in agriculture should

include the added burden of ground and surface water contamination by the toxic

metabolites CDA and HDA.

D. Cross-Reactivity of CDA and HDA Mutagens in Alachlor Immunoassays

Immunoassay kits are currently available for the detection of alachlor in ground and

surface waters. Alachlor residue levels measured by these kits show a slight positive bias

when compared to chromatographic analysis of the same samples. This indicates a

degree of cross-reactivity to structurally-related compounds, notably alachlor soil

metabolites and, in particular, the ethanesulfinic acid (ESA) metabolite (Baker et al.

1993). It was determined that CDA and HDA are not cross reactive in three

commercially available immunoassay kits (i.e., Ohmicron, Agri-Diagnostics, and

Millipore; Tessier 1994). Therefore, these toxicologically significant metabolites are not

included in the quantitation of alachlor contamination via monitoring schemes based on

currently available immunoassays.

11 E. Utility of an Alachlor Metabolite Immunoassay

CDA and HDA have been detected in ground and surface waters in the course of monitoring for alachlor and other major use herbicides. However, there are relatively little data on the incidence of these compounds in other environmental compartments, or the overall extent of their incidence in the environment. Additionally, there are virtually no data concerning the rate of formation, movement, persistence, and further degradation of these mutagens. In addition to the overall impact that CDA and HDA may have on organisms exposed to these mutagens in soils and water, there is the additional potential for chronic human exposure from contaminated drinking water supplies and via bioaccumilation.

The development of immunoassay technology specifically directed at the detection of

CDA and HDA was deemed meritorious in that it would provide a sensitive, fast, inexpensive means of measuring the incidence of these known mutagens in environmental matrices. This in turn would provide a better understanding of their environmental fate and whether operational factors, such as formulation type or soil management practices, mitigate or enhance their formation. Ultimately, this will clarify how the environmental degradation of alachlor occurs and which environmental factors result in the increased formation of mutagenic metabolites. Information on these processes will allow the development of strategies that will attenuate the overall risks posed by continued use of this carcinogenic herbicide.

To date, the low level detection of CDA and HDA in water samples has been accomplished by gas chromatography with ion-trap mass spectrometry (Pereira et al.

12 1990). While this methodology allows determination of part per trillion level residues in water samples, the cost of instrumentation and the skill level required to operate it is prohibitive for significant, large-scale monitoring programs. Immunoassays specific for

CDA and HDA would compliment this chromatographic analysis by providing the ability to screen large numbers of samples quickly and inexpensively. More involved analytical procedures can thus be performed as confirmation of a much smaller set of positive samples.

A new direction in environmental analysis exploits the specific binding properties of antibodies as immunosorbents in sample pretreatment (Henion et al. 1997). Antibodies bound to a solid support can be used as a solid-phase extraction mechanism. Therefore, analyte-specific or chemical class-specific concentration and cleanup of samples for traditional instrumental analyses is possible. The availability of CDA and HDA specific antibodies will make this new analytical strategy possible for these toxicologically significant alachlor metabolites.

13 CHAPTER II

SYNTHESIS OF HAPTENIC DERIVATIVES FOR THE

PREPARATION OF CDA AND HDA IMMUNOGENS

A. Chemicals

2,6-Diethylaniline, ethyl chloroacetate, chloroacetyl chloride, 4-ethylphenol, aluminum chloride and sodium nitrite were obtained from Aldrich Chemical Co.

(Milwaukee, WI). N-hydroxysuccinimide (NHS), N,N’-dicyclohexylcarbodiimide

(DCC), N-acetyl homocysteine thiolactone (AHT), ethyl 4-bromobutyrate, ethyl 5- bromovalerate, bovine serum albumen (BSA), human serum albumen (HSA) and ovalbumin (OVA) were obtained from Sigma (St. Louis, MO). Goat anti-rabbit IgG conjugated to horseradish peroxidase was obtained from Jackson ImmunoResearch

Laboratories (West Grove, PA). 2-Chloro-2’,6’-diethylacetanilide (CDA, 99%) and 2- hydroxy-2 ,6 -diethylacetanilide (HDA, 99%) were synthesized in the Mass Spectrometry

Laboratory, College of Food and Natural Resources, University of Massachusetts /

Amherst (Potter and Carpenter 1995).

B. Apparatus

Infrared (IR) analysis was conducted on a Perkin-Elmer 1330 spectrophotometer

(Norwalk, CT) scanning from 4000 to 200 cm1. Proton nuclear magnetic resonance (*H

NMR) analysis was conducted on a Bruker / IBM 200AC (Karlsruhe, Germany) spectrometer operating at 200 MHz. Chemical shifts in ppm are reported compared to a tetramethylsilane (TMS) standard. Gas chromatographic / mass spectral (GC/MS)

14 analysis was conducted on a Hewlett-Packard 5890 II gas chromatograph / 5971 mass selective detector (Palo Alto, CA) equipped with a 30 meter HP-5 column, 0.25 pm film, temperature programmed from 70 °C (1 min) to 275 °C (1 min) at 15 C/min. Injector temperature was 250 °C, the interface 280 °C, and the mass range scanned from 35 - 550 mass units. Immunoassays were determined on a Softmax™ Plate Reader (Molecular

Devices, Sunnyvale, CA).

C. Synthesis of Haptens

1. Synthesis of Hapten C2-CDA

The synthesis of a haptenic derivative of CDA containing a 2-carbon spacer group attached to the nitrogen moiety (C2-CDA) for subsequent linkage to carrier protein molecules is outlined schematically in Figure 3. Details of the synthetic procedures and structural confirmation are described below, a. 2,6-Diethylaniline-N-ethyl acetate (2)

2,6-Diethylaniline (1, 7.5 g; 0.05 mol) and ethylchloroacetate (12.3 g; 0.1 mol) were heated with stirring in the presence of Na2C03 (4.2 g; 0.05 mol) for 4 h at 140 °C. The reaction mixture was removed from heat and distilled water (dH20; 20 ml) added. Both aqueous and organic phases were transferred to a separatory funnel and ethyl acetate (15 ml) added. The crude reaction mixture was extracted by shaking for 1 min and the aqueous phase discarded. The organic phase was washed twice with 1 M HC1 (25 ml) and dried over Na2S04. Ethyl acetate was removed from the organic phase by rotory evaporation to yield 11.44 g crude 2. The crude product was purified by dry column chromatography (63 - 200 pm silica, 40% ethyl acetate : hexane) to yield an

15 Figure 3. Synthesis of hapten C2-CDA (4) containing a 2-carbon spacer group attached to the nitrogen moiety.

16 H\ / H

i. ethylchloroacetate ii. 0.7 M NaOH iii. chloroacetyl chloride

17 amber oil (90%). IR (neat, cm'1): 3380 (N-H), 2960-2865 (alkyl C-H), 1740 (C-O),

1455 (aryl C=C), 1370-1340 (2° aryl C-N), 1205 (ester C-O-C), 860,750 (1,2,3 trisub.

aryl C-H). 'H NMR (200 Hz, CDC13): 8 1.2 (t, 9H, -CH3), 2.7 (q, 4H, -CH2-), 3.8 (s, 2H,

N-CH2-), 4.0 (s, 1H, N-H), 4.2 (q, 2H, -COOCH2-), 7.0 (m, 3H, aryl -CH). GC/MS: tr =

8.9 min, m/z = 235 (M+), 162, 147. b. 2,6-Diethylaniline-N-acetic acid (3)

2.6- Diethylaniline-N-ethyl acetate (2, 1.0 g) was dissolved in 40 ml 0.7 M NaOH / acetone (65:35 v:v). The reaction mixture was stirred at ambient temperature for 14 h.

Acetone was removed under vacuum and the aqueous mixture acidified (H3PO4) to pH 3-

4. Crude 3 was extacted with 3 x 20 ml ethyl actetate. The combined solvent extracts

were dried over Na2S04 and the solvent removed under vacuum. The product was

purified by dry column chromatography (63 - 200 pm silica, 25 % ethyl acetate : hexane)

to yield an amber oil (80%). IR (neat, cm"1): 3380 (N-H), 3300 - 2500 (acid O-H), 1730

(acid C=0), 1450 (aryl C=C), 1400 - 1380 (2° aryl C-N), 1220 (broad, -COOH), 805, 730

(1,2,3 trisub. aryl C-H). GC/MS (methyl ester): tr = 8.3 min, m/z = 221 (M+), 162, 147,

132.

c. 2-Chloro-2’,6’-diethyl(N-acetic acid)acetanilide (4)

2.6- Diethylaniline-N-acetic acid (3, 1.5 g) was dissolved in 45 ml methylene chloride.

Chloroacetyl chloride (1 g) in methylene chloride (15 ml) was added dropwise with

stirring (CAUTION!) to the solution of 3. The reaction continued for 2 h at ambient

temperature with constant stirring. The reaction mixture was washed with 2 x 25 ml

dH20, and the organic layer dried over Na2S04. The solvent was removed under vacuum

to yield 1.4 g of a purple oil. The product was purified by dry-column chromatography

18 (63 - 200 pm silica, 25 % ethyl acetate : hexane) to yield an amber oil (84%). IR (neat, cm’1): 3005-2860 (C-H), 3300-2500 (br, O-H), 1735 (CO), 1680 (3° amide C=0), 1450

(aryl CO), 1430-1380 (3° aryl amine C-N), 1250-1200 (br, -COOH), 800-760 (1,2,3 trisub. aryl C-H). GC/MS (methyl ester): tr = 11.6 min, m/z = 297 (M+), 248, 188, 160.

2. Synthesis of Hapten C4-CDA

The synthesis of a haptenic derivative of CDA containing a 4-carbon spacer group attached to the nitrogen moiety (C4-CDA) for subsequent linkage to carrier protein molecules is outlined schematically in Figure 4. Details of the synthetic procedures and structural confirmation are described below, a. 2,6-Diethylaniline-N-ethylbutyrate (5)

2,6-Diethylaniline (1, 7.5 g, 0.05 mole) and ethyl 4-bromobutyrate (19.5 g, 0.1 mole) were stirred at 140 °C for 4 h in the presence of sodium bicarbonate (5.3 g, 0.05 mole).

The reaction mixture was removed from heat and dH20 (20 ml) added. Both aqueous and organic phases were transferred to a separatory funnel and ethyl acetate (15 ml) added.

The crude reaction mixture was extracted by shaking for 2 min and the aqueous phase discarded. The organic phase was washed twice with 1 M HC1 (20 ml), and dried over

Na2S04. Ethyl acetate was removed from the organic phase by rotory evaporation to yield 14 g of a purple-black oil. The crude product was purified by dry column chromatography (63 - 200 pm silica, 10% ethyl acetate : hexane) to yield a pale yellow oil

(90%). IR (neat, cm'1): 2950-2850 (alkyl C-H), 1750 (C=0), 1450 (aryl C=C), 1360 (2° aryl C-N), 1250 (ester C-O-C). GC/MS: tr = 11.5 min, m/z = 263 (M+), 218, 162, 148,

132.

19 Figure 4. Synthesis of hapten C4-CDA (7) containing a 4-carbon spacer group attached to the nitrogen moiety.

20 H, H

7 i. ethyl 4-bromobutyrate ii. 0.7 M NaOH iii. chloroacetyl chloride

21 b. 2,6-Diethylaninline-N-butyric acid (6)

2.6- Diethylaniline-N-butyrate (5, 1.0 g) was dissolved in 40 ml 0.7 M NaOH / acetone

(65:35 v:v). The reaction mixture was stirred at ambient temperature for 18 h. Acetone

was removed under vacuum and the aqueous mixture acidified (H3PO4) to pH 3-4. Crude

3 was extacted with 3 x 20 ml ethyl actetate. The combined extracts were dried over

Na2S04 and the solvent removed under vacuum to yield 0.75 g of a yellow oil.

c. 2-Chloro-2’,6’-diethyl(N-butyric acid)acetanilide (7)

2.6- Diethylaniline-N-butyric acid (6, 0.75 g) was dissolved in 25 ml methylene

chloride. Chloroacetyl chloride (0.9 g) in methylene chloride (6 ml) was added dropwise

with stirring (CAUTION!) to the solution of 6. The reaction continued for 2 h at ambient temperature with constant stirring. The reaction mixture was washed with 3 x 25 ml dH20, and the organic layer dried over Na2S04. The solvent was removed under vacuum to yield 0.75 g of a purple oil that solidified on standing. The product was purified by dry-column chromatography (63 - 200 pm silica, 30 % ethyl acetate : hexane) to yield 0.6 g white crystals.

3. Synthesis of Hanten Phe-CDA

The synthesis of a haptenic derivative of CDA containing a five-carbon spacer group attached to the aromatic ring (Phe-CDA) for subsequent linkage to carrier protein molecules is outlined schematically in Figures 5 and 6. Details of the synthetic procedures and structural confirmation are described below, a. 3,5-Diethylphenol (9)

4-Ethylphenol (8, 50g; 0.41 mole) and aluminum chloride (125g; 0.94 mole) were

22 Figure 5. Synthesis of 2,6-diethyl-4-hydroxyaniline.

23 i. aluminum chloride /110°C/4h ii. sodium nitrite iii. H2 / Pd catalyst

24 Figure 6. Synthesis of hapten Phe-CDA (14) containing a 5-carbon spacer group attached to the aromatic ring.

25 OH 11

9 H CCH2C1 N

111

i. ethyl 4-bromobutyrate ii. chloroacetyl chloride iii. 0.7 MNaOH

26 heated with stirring at 110 °C + 5 °C for 4 h. The reaction mixture was cooled to 60 C,

poured onto crushed ice and steam distilled to yield 41g of a crude phenolic mixture

containing phenol, 3,5-diethylphenol and by-products. The phenolic mixture was

fractionally distilled under vacuum (15 mm Hg). 3,5-Diethyl-

phenol, collected as the remaining fraction after removal of phenol and related by¬

products, was recrystallized from petroleum ether to yield 3.8g of white, needle crytals

with a distinct phenolic odor. Melting point, 77 °C. Purity, 99.9% (GC/MS). IR (KBr

pellet, cm'1): 3320-3180 (s,b, O-H ); 2960, 2940, 2880 (s, CH2, CH3); 1600, 1450 (Ar

ring). GC/MS: tr = 10.27 min, m/z - 150 (M+), 135 (M-CH3), 121 (M-C2H5).

b. 3,5-Diethyl-4-nitrosophenol (10)

3.5- Diethylphenol (9, 3.6g, 24 mmol) was dissolved in reagent ethanol (25 ml) and

concentrated HC1 (25 ml) was added with stirring. The reaction mixture was cooled to 0

°C and NaN03 (2.5 g, 30 mmol), dissolved in 3 ml dH20, was added dropwise. The

reaction was continued for 2 h at 10 °C. The reaction mixture was poured onto ice-water

and filtered to yield 3.4 g of rust-colored crystals. Melting point, 126 °C (decomp.). IR

(KBr pellet, cm'1): 3460-2600 (s,b, O-H and -CH2-, -CH3); 1720, 1685, 1655 (Ar ring),

1500 (s, -N=0), 1140 (s, C-OH), 1000 (C-N). c. 2,6-Diethyl-4-hydroxyaniline (11)

3.5- Diethyl-4-nitrosophenol (10, 3.4 g in 40 ml THF) was hydrogenated at atmospheric pressure in the presence of palladium catalyst (380 mg). [CAUTION: The palladium catalyst can ignite solvent vapors. The hydrogenation apparatus must be flushed with inert gas prior to addition of catalyst.] The hydrogenation reaction continued for 3 h, after which the reaction solution was filtered through Celite 545 to remove the catalyst.

27 Removal of the solvent under vacuum yielded 3 g of orange crystals. Melting point, 118

°C. Purity, 97.6% (GC/MS). GC/MS: tr = 8.4 min, m/z = 165 (M+), 150 (M-CH3), 135.

Elemental analysis (theoretical values in parentheses): C: 72.38% (72.73); H: 9.3% (9.1);

N: 8.36% (8.5); O: 9.96% (9.7). d. 2,6-Diethyl-4-[4’-(ethoxycarbonyl)butoxy]aniline (12)

2.6- Diethyl-4-hydroxyaniline (11, 1 g, 6 mmol) was dissolved in 10 ml dimethyl sulfoxide. The stirred solution was cooled to 15 °C and potassium hydroxide (0.5 g, 9 mmol) added. The solution was returned to ambient temperature and ethyl-5 - bromovalerate (1.9 g, 9 mmol) was slowly added over the course of 70 min, after which the reaction was stopped by pouring the solution into 10 ml 4 MHC1 at 0 °C. The acidic solution was extracted with diethyl ether (3 x 50 ml) to remove unreacted ethyl-5 - bromovalerate, and then basified with 15 ml 2 MNaOH. The basic solution was extracted again with diethyl ether (3x30 ml), the combined ethereal extracts were washed with distilled water (2 x 50 ml), brine (2 x 50 ml) and dried over Na2SC>4. The solvent was removed under nitrogen to yield 0.3 g of a brown oil. Purity, 99.7%

(GC/MS). IR (neat, cm'1): 2950 (-CH2-, -CH3), 1725 (C=0, ester), 1590 (-N-H), 1470 (-

CH2-). GC/MS: tr= 14.14 min, m/z = 293 (M+), 248, 165/164, 129, 101. e. 2-Chloro-2’,6’-diethyl-4’-[4”-(ethoxycarbonyl)butoxy]acetanilide (13)

2.6- Diethyl-4-[4’-(ethoxycarbonyl)butoxy]aniline (12, 0.3 g, 1 mmol) was dissolved in

10 ml methylene chloride. Chloroacetyl chloride (170 mg, 1.5 mmol, diluted in 1.5 ml methylene chloride) was added at room temperature to the stirred solution of 12 in 100 pi aliquots at 2 min intervals. The reaction was continued 30 min after the last chloroacetyl chloride addition. The reaction mixture was washed with water (3 x 20 ml), followed by

28 brine (3 x 20 ml), dried over Na2S04 and the solvent removed to yield 0.33 g of a brown solid. Purity 99.9% (GC/MS). IR (Nujol, cm’1): 3240, 2940, 1740 (s, ester C=0), 1660,

1600. GC/MS: tr= 15.00 min, m/z = 369 (M+), 129, 101 (M - .C2H5CO.OC2H5).

Elemental analysis (theoretical values in parentheses): C: 61.55% (61.7), H: 7.57%

(7.63), N: 3.68% (3.79), Cl: 9.38% (9.58), O: 17.82 (17.30). f. 2-Chloro-2’,6’-diethyl-4’-[4”(carboxylic acid)butoxy]acetanilide (14)

2-Chloro -2’,6’-diethyl-4’-[4’’-(ethoxycarbonyl)butoxy]acetanilide (13, 163 mg, 0.4 mmol), dissolved in 1 ml acetone, was stirred in the presence of 2 MKOH (1.5 ml) at ambient temperature for 24 h. The reaction solution was neutralized with 4 M HC1 and extracted with diethyl ether (3x7 ml). The combined extracts were washed with water (2 x 10 ml), brine (2x10 ml) and dried over Na2SC>4. The solvent was removed under vacuum to yield 152 mg of a brown semi-solid. Crude 14 was purified by thin-layer chromatography (silica gel, 2000 p, ethyl acetate : hexane : acetic acid, 25:70:5) to yield

103 mg white crystals. Melting point, 132 °C (decomp). Purity, 99% (TLC). IR (KBr pellet, cm'1): 3680-2500 (m, b, acidic -O-H), 3285 (s, amide N-H), 1720 (s, acidic C=0),

1660. GC/MS (methyl ester): tr = 10.9, m/z = 355 (M+), 192, 115.

D. Preparation of Hapten-Protein Conjugates

1. Preparation of Active Esters of the CPA Haptens

The acid moieties of C2-CDA, C4-CDA and Phe-CDA were modified to activated esters to allow spontaneous coupling of the haptens to amine residues on the carrier proteins (Fig. 7, Tijssen 1985).

29 Figure 7. Formation of active esters of C2-CDA, C4-CDA and Phe-CDA via N-

hydroxysuccinimide.

30 Hapten R 9 ,CCH2C1 C2-CDA

C4-CDA

9 H ,CCH2CI Phe-CDA N

CH2CH2CH2CHP i. N-hydroxysuccinimide ii. N,N'-dicyclohexylcarbodiimide iii. activated ester of hapten iv. N,N'-dicyclohexylurea

31 a. General method for the formation of active esters

C2-CDA, C4-CDA or Phe-CDA (0.1 - 0.6 g) and N-hydroxysuccinimide (0.1 - 0.26 g) were dissolved in 5 - 10 ml THF. The solution was cooled to 0 °C and N,N - dicyclohexylcarbodiimide (50 - 500 mg) was added with stirring. The reaction was continued at -20 °C for 24 h. The cold solution was filtered to remove most of the dicyclohexyl urea (DCU) that was formed during the reaction. Methylene chloride was added to the filtrate, which was washed with 2 x 10 ml dT^O followed by 2 x 10 ml saturated NaCl solution. Solvent was removed under vacuum from the dried organic layer (Na2S04) to yield an amber, oily-crystalline residue consisting of the activated ester and residual DCU. The esters were purified by thin-layer chromatography (20 x 20 cm,

2000 pm silica, 25 % ethyl acetate : hexane). b. Structural confirmation of the active ester of C2-CDA

IR (neat, cm"1): 3320 (w, imide N-H), 2970-2880 (m, alkyl C-H), 1830, 1795, 1750,

1690 (s, aliphatic, imide, 3° amide C=0), 1460-1370 (m, 3° arylamine C-N), 1240 - 1200

(s, -CO.OR), 810 (m, 1,2,3 trisub aryl C-H). 'H NMR (200 Hz, CDC13): 5 1.2 (t, 6H, -

CH3), 2.6 (q, 4H, alkyl -CH2), 2.7 (s, 4H, imide -CH2), 3.7 (s, 2H,-N-CH2-COOR), 4.5 (s,

2H, Cl-CH2-CO.R), 7.2 (m, 3H, aryl CH). MS: m/z= 380 (M+), 266, 238, 216, 188, 162.

Elemental analysis (theoretical values in parentheses): C: 57.13% (56.77%); H: 5.85%

(5.56%); N: 7.0% (7.36%); O: 21.08% (21.00%); Cl: 8.93% (9.30%). c. Structural confirmation of the active ester of C4-CDA

IR (neat, cm"1): 3400 - 3300 (w, imide N-H), 2940 (m, alkyl C-H), 1795, 1760, 1740,

1650 (s, aliphatic, imide, 3° amide C=0), 1460-1320 (m, 3° arylamine C-N), 1200 (s, -

CO.OR), 1050, 850 (m, 1,2,3 trisub aryl C-H). Elemental analysis (theoretical values in

32 parentheses): C: 59.58% (59.04%); H: 6.43% (5.70%); N: 6.69% (6.89%), O. 18.85 yo

(19.66%); Cl: 8.45% (8.71%). d. Structural confirmation of the active ester of Phe-CDA

IR (neat, cm’1): 3290 (s, imide N-H), 2940 (s, alkyl C-H), 1795, 1760, 1670, 1630

(aliphatic, imide, 3° amide C=0), 1210 (s, -CO.OR), 1170, 1070 (m, 1,2,3 trisub aryl C-

H), 870.

2. Conjugation of CPA Haptens to Carrier Proteins a. Conjugation of active esters

The activated esters of C2-CDA, C4-CDA and Phe-CDA, dissolved in THF, were stirred with protein (BSA, HSA or OVA) in phosphate -buffered saline (PBS) at 0 °C, at a 50:1 molar ratio (Table 2). The reaction was continued for 48 h at 4 °C. The conjugate solutions were dialyzed against dTCO (8-10 changes) at 4 °C and then lyophilized. b. Conjugation via N-acetyl homocysteinethiolactone

CDA was also directly coupled to protein carrier molecules via N-acetyl homocysteine¬ thiolactone [AHT, Fig. 8, (Feng et al. 1992)]. The AHT thiolation of free amine groups on lysine residues of the protein carrier molecules provided a high density of reactive sites. Under basic conditions, the thiol groups spontaneously formed a covalent linkage with the chloroacetanilide moiety of CDA. Thus, BSA or OVA (200 mg) was dissolved in 6 ml dFfO. AHT (18.4 mg, 115 mmol) was added to the protein with stirring and the solution chilled to 0 °C. CDA dissolved in 1 ml dioxane was added dropwise and the reaction mixture adjusted to pH > 10 with carbonate - bicarbonate buffer (1.0 M, pH=l 1).

After stirring at 0 C for an additional 15 min, the solution was transferred to a 50 °C

33 water bath and the reaction continued for 2 h. The reaction solution was neutralized with

phosphate buffer (1.0 M, pH=7.2) and dialyzed against df^O (8 changes). The BSA-

AHT-CDA conjugate was used for antibody production and the OVA-AHT-CDA

conjugate was used as the ELISA plate-coating antigen.

Hapten densities were determined by titrating free amine moieties with trinitrobenzene

sulfonic acid (TNBS, Habeeb 1965, Table 2). The hapten - protein linkages for the four

immunogens are shown in Figure 9.

Table 2. Hapten Densities of Immunogens and Plate-Coating Antigens

Hapten-Protein Conjugate Hapten Density1

Immunogens

BSA2-C2-CDA 78.4%

BSA-C4-CDA 83.8%

BSA-Phe-CDA 63.2%

BSA-AHT-CDA 46.8% Plate-Coating Antigens

HSA3-C2-CDA 66.9%

BSA-C4-CDA 83.8%

OVA4 5-Phe-CDA 16.8%

OVA-AHT-CDA __5

1. Hapten densities reported as percent of available lysine residues (free amines) BSA = 61; OVA = 20. 2. Bovine serum albumin. 3. Human serum albumin. 4. Ovalbumin (chicken egg albumin). 5. Unable to determine due to solubility problems.

34 Figure 8. Direct coupling of CD A to protein carrier molecules via N-acetyl homocysteine thoilactone (AHT). The protein was bovine serum albumin (BSA) for the immunizing

conjugate and ovalbumin (OVA) for the plate-coating conjugate.

35 o

CDA

NH2-Protein T 2

36 37 BSA-Phe-CDA BSA-AHT-CDA

38 CHAPTER III

DEVELOPMENT OF A COMPETITION ENZYME-LINKED IMMUNOSORBENT

ASSAY (cELISA) FOR THE DETECTION OF CD A AND HD A

A. Antibody Production

Immunizations and blood collections were performed by Animal Care and Research

Services, Graduate School, University of Massachusetts, Amherst. Standard immunization protocols were followed (Coligan et al. 1995). Female white New Zealand rabbits (2-3 kg) were initially immunized with 250 jug hapten-protein conjugate solublized in 250 pi PBS (pH 7.2) and emulsified 1:1 with Freund’s complete adjuvant

(Table 3). Groups of three rabbits were immunized with BSA-C2-CDA and BSA-C4-

CDA, and two pairs of rabbits immunized with BSA-Phe-CDA and BSA-AHT-CDA

(Table 4). Subsequent boosts of 250 pg hapten-protein conjugate in PBS, emulsified with

Freund’s incomplete adjuvant, were administered at three week intervals. All injections were administered intradermally at multiple locations on the back. Approximately 10 ml blood was collected 7 to 10 days following each immunization. Serum was collected by centrifugation (i.e., 3500 rpm, 15 min, 4 C) and diluted 1:1 with glycerol. Thimerosal was added at 0.01% as a preservative and the sera were stored at -20 °C.

B. Procedures for Enzyme Immunoassays

Immunoassays were conducted on 96-well microtiter plates (Nunc). Plates were coated with hapten-protein conjugate in carbonate / bicarbonate buffer (0.05 M, pH 9.6,

39 Table 3. Hapten-Carrier Conjugates Used for Immunization and Plate Coating

cELISA # Plate-Coating Antigen

1 BSA'-C2-CDA HSA2-C2-CDA

2 BSA-C4-CDA BSA-C4-CDA

3 BSA-Phe-CDA OVA3-Phe-CDA

4 bsa-aht4-cda OVA-AHT-CDA

1. Bovine serum albumin. 2. Human serum albumin. 3. Ovalbumin (chicken egg albumin). 4. N-acetyl homocysteine thiolactone.

Table 4. Antisera Designations

Immunogen Rabbit Number Serum Pool Number

BSA-C2-CDA 1 RAX1 2 RAX2 41 RAX4 BSA-C4-CDA 5 RAX5 6 RAX6 7 RAX7 BSA-Phe-CDA 8 RAX8 9 RAX9 BSA-AHT-CDA 10 RAX10 11 RAX11

1. Rabbit # 3 died during initial immunization.

50 jul/well) by incubating overnight at room temperature. After washing, plates were blocked with protein (0.25% OVA or BSA in PBS ammended with 0.05% Tween-20

(PBST)). Dried plates were stored in zip-lock bags at 4 °C.

The antibody capture format used in all assays is shown in Figure 10. In this format,

40 41

U Ou X l OW)

-C -D rt i n < < rt

OJ C3 CD 3 O "c* W) o i—■* U ^ 53c "i_ bo ^ v<

42 analyte in solution competes with plate-bound antigen for CDA-specific antibody (i.e., primary antibody) binding. Maximum binding of antibody to plate-bound antigen occurs in the absence of analyte. The presence of analyte inhibits primary antibody binding to antigen in proportion to analyte concentration. Plate-bound antibodies are quantitated by addition of secondary antibody specific for primary antibody. The secondary antibody is conjugated to the reporter enzyme, horseradish peroxidase (HRP). Visualization of bound antibodies is accomplished with a tetramethylbenzidine / peroxide (TMB-H2O2) substrate system, which forms a colored product that is quantified spectrophotometrically. In this format, optical density (OD) is inversely proportional to analyte concentration. ODs were reported directly versus analyte concentration, or as percent ratios of control values (i.e., water blank) versus analyte concentration. The percent ratios %B/Bo were determined by the formula:

%B/Bo = (ODsampie/ODControl)xlOO.

The IC50, or concentration of analyte resulting in 50% inhibition of antibody binding, was extrapolated from plots of analyte concentration versus %B/Bo.

Optimal concentrations of antisera, plate coating antigen and secondary reporter antibodies were determined by checkerboard assays utilizing standard procedures as outlined in Coligan et al. (1995) and Harlow and Lane (1988). Antisera diluted in PBST was added to the microtiter plate (50 pl/well) and incubated at 37 °C for 60 min. After washing, bound antibodies were quantified using goat anti-rabbit IgG antibodies conjugated to HRP (50 pl/well), incubated for 30 min at 37 °C. The TMB-H202 for HRP was prepared by diluting 200 pi TMB stock solution (25 mmol in DMSO) and 17 pi

H202 (3% solution) in 12.5 ml sodium acetate buffer (0.1 M, pH 5.5). TMB- H202 (100

43 jul/well) was incubated for 10 to 25 min at room temperature and the reaction stopped by

adding 50 pl/well 4 MH2SO4. OD was determined at 450 nm. In competition assays,

solutions of CDA, HDA and cross-reactivity test compounds (from methanol or acetone

stock solutions) were diluted in dE^O. Competitor solutions were added immediately prior to addition of primary antisera dilutions. The limit of detection was determined as the concentration that resulted in an OD value less the dH^O control mean OD value minus 3 standard deviations (Brady 1995).

C. Competition Enzyme-Linked Immunosorbent Assays (cELISA)

1. cELISA 1: Anti-BSA-C2-CDA Antisera a. Checkerboard assays

Microtiter plates were coated with HSA-C2-CDA at 60, 125, 250, and 500 ng/well.

Antisera RAX1, RAX2 and RAX4 were serially diluted 1:125, 1:500, 1:2000, 1:8000,

1:32,000 in PBST. Goat anti-rabbit IgG-HRP was serially diluted 1:1250, 1:5000,

1:10,000, 1:20,000, 1:40,000, 1:80,000 in PBST. Assays were run as described in section

III.B. Results from a typical assay are presented in Table 5. b. Competition assays

Standard solutions of CDA in dL^O were added to the plate prior to addition of primary antisera dilutions. Based on the results of the checkerboard assays, antisera RAX1,

RAX2 and RAX3 were diluted 1:4000 in PBST. Goat anti-rabbit IgG-HRP was diluted

1:40,000 in PBST. Binding of RAX1 antibodies was inhibited by CDA (Fig. 11) It was necessary, however, to incubate the initial competition step for 16 h in order to detect

44 Table 5. Checkerboard Assay for Antisera RAX1 versus BSA-C2-CDA and HSA- C2-CDA Plate-Coating Antigens

BSA-C2-CDA|HSA-C2-CDA

1° Ab: 1:125 1:500 1:2K 1:8K 1:32K Ctrl1 1:125 1:500 1:2K 1:8K 1:32K Ctrl

2° Ab 1:1250 2.3492 1.827 1.366 0.996 0.552 0.292 2.936 2.646 2.067 1.738 1.116 0.357

1:2500 2.241 1.619 1.131 0.654 0.408 0.270 2.640 2.288 1.866 1.734 0.730 0.449

1:5000 1.878 1.409 0.766 0.469 0.317 0.188 2.450 2.178 1.775 1.014 0.498 0.307

1;10K 1.458 0.941 0.507 0.274 0.204 0.141 2.253 2.055 1.129 0.668 0.299 0.266

1:20K 1.356 0.617 0.410 0.180 0.107 0.081 1.936 1.512 0.730 0.440 0.255 0.199

1:40K 0.850 0.448 0.243 0.115 0.075 0.067 1.597 0.947 0.498 0.203 0.159 0.103

1:80K 0.510 0.365 0.174 0.094 0.069 0.064 1.186 0.746 0.374 0.211 0.147 0.100

Ctrl 0.056 0.054 0.055 0.062 0.058 0.054 0.079 0.085 0.085 0.091 0.082 0.063

1. Control (buffer only). 2. OD determined at 450 nm.

inhibition of antibody binding by CD A. The concentration range for inhibition was 1.0 to

100 jag/ml CDA, which indicated that RAX1 antibodies had only slight recognition of

CDA. RAX2 and RAX3 were not inhibited by CDA (i.e., did not recognize CDA) and were not further characterized, c. Cross-reactivity to related compounds

RAX1 showed far greater affinity to alachlor than to the intended target analyte CDA

(Fig. 11). The detection range was 0.05 to 100 jug/ml alachlor. Thus, BSA-C2-CDA did not result in the production of CDA- specific antibodies.

45 Figure 11. Competitive inhibition of antiserum RAX1 by CDA and alachlor. Microtiter plates were coated with 60 ng/well HSA-C2-CDA. Antiserum RAX1 was diluted

1:20,000 in PBST. Goat anti-rabbit IgG was diluted 1:10,000 in PBST. Absorbance values (OD) are the mean of eight replicates. Error bars represent ± 1 SD.

46 1.0

d o

0.01 0.10 1.00 10.00 100.00 Concentration (ug/ml) 2. cELISA 2: Anti-BSA-C4-CDA Antisera a. Checkerboard assays

Microtiter plates were coated with BSA-C4-CDA at 60, 125, 250, and 500 ng/well.

Antisera RAX5, RAX6 and RAX7 were serially diluted 1:1,000, 1:5,000, 1:10,000 and

1:20,000 in PBST. Goat anti-rabbit IgG-HRP was serially diluted 1:5,000, 1:10,000,

1:20,000 and 1:40,000 in PBST. Assays were run as described in section III.B. Typical results are shown in Table 6. b. Competition assays

Standard solutions of CDA and alachlor in dH^O were added to the plate prior to addition of primary antisera dilutions. Based on the results of the checkerboard assays, antisera RAX5, RAX6 and RAX7 were diluted 1:2,000 in PBST. Goat anti-rabbit IgG-

HRP was diluted 1:5,000 in PBST. Binding of RAX7 was only slightly inhibited by

CDA, which resulted in an IC50 value greater than 100 jag/ml (Fig. 12). Alachlor, however, showed greater inhibition with an IC50 value of 3 jug/ml (Fig. 12). Binding of

RAX5 and RAX6 showed a similar affinity for alachlor but not for CDA and were not further characterized. c. Cross-reactivity to related compounds

The inhibition of RAX7 antiserum by the chloroacetanilide herbicides acetochlor, butachlor, metolachlor and propachlor (Fig. 13) was determined in competition assays over the concentration range 0.05 to 100 pg/ml. Although RAX7 responded to alachlor, cross reactivity to these structurally-related compounds was negligible (Table 7). Thus,

BSA-C4-CDA also did not result in the production of CDA-specific antibodies.

48 Table 6. Checkerboard Assay for the Detection of BSA-C4-CDA by Antisera RAX5, RAX6 and RAX7

1° Ab: RAX 5_ RAX6_RAX 7_

1° Ab Dilution VAK U5K 1:10K 1:20K 1:1K 1:5K 1:1QK 1:20K 1:1K 1:5K 1.10K 1:2QK

2° Ab 1:5K 1.0201 0.829 0.701 1.159 1.090 0.889 1.524 1.414 1.223 1.535 1.448 1.218

1.000 0.815 0.646 1.103 1.020 0.851 1.475 1.282 1.186 1.446 1.402 1.310

2° Ab 1:10K 0.837 0.734 0.576 1.171 0.955 0.763 1.394 1.309 1.091 1.460 1.380 1.187

0.056 0.047 0.041 0.044 0.042 0.042 0.044 0.042 0.042 0.044 0.043 0.048

2° Ab 1:20K 0.969 0.776 0.612 1.203 1.070 0.733 1.504 1.354 1.212 1.540 1.426 1.292

0.736 0.631 0.435 1.120 0.900 0.698 1.425 1.195 1.074 1.552 1.365 1.228

2° Ab 1:40K 0.755 0.597 0.426 0.963 0.842 0.611 1.312 1.138 0.955 1.399 1.214 1.072

0.056 0.049 0.044 0.052 0.052 0.052 0.053 0.058 0.048 0.056 0.053 0.047

1. OD determined at 450 nm.

Table 7. Cross-Reactivity (Reported as IC50) of Chloroacetanilide Herbicides to Antiserum RAX7

Alachlor 3 jag/ml Metolachlor >100 jag/ml

Acetochlor >100 jag/ml Propachlor >100 jag/ml

Butachlor >100 jag/ml

49 Figure 12. Competitive inhibition of antiserum RAX7 by CDA and alachlor. Microtiter plates were coated with 60 ng/well BSA-C4-CDA. Antiserum RAX7 was diluted

1:20,000 in PBST. Goat anti-rabbit IgG was diluted 1:5,000 in PBST. Absorbance values (OD) are the mean of eight replicates. Error bars represent + 1 SD.

50 1.0

0.8

0.6 d o

0.4

0.2

51 Figure 13. Structures of chloroacetanilide herbicides and related compounds tested for cross-reactivity. I, alachlor; II, hydroxyalachlor; III, alachlor ethanesufonic acid (alachlor-

ESA); IV, 2,6-diethylaniline; V, acetochlor; VI, butachlor; VII, metolachlor;

VIII, propachlor.

52 o 9 CHjOCH^ xch2oh CH3OCH2^ XCH2C1 N

53 3. cFT.ISA 3: Anti-BSA-Phe-CDA Antisera a. Checkerboard assays

Microtiter plates were coated with OVA-Phe-CDA at 0.5, 1.0, 5.0, and 10.0 ng/well.

Antisera RAX10 and RAX 11 were serially diluted 1:5000, 1:10,000 and 1:20,000 in

PBST. Goat anti-rabbit IgG-HRP was serially diluted 1:5,000, 1:10,000 and 1:20,000 in

PBST. Assays were run as described in section III.B. Results are shown in Table 8. b. Competition assays

Initial competition assays showed that RAX11 but not RAX 10 specifically recognized

CD A in solution. RAX11 was used in all subsequent competition assays (Fig. 14). A stock solution of CDA in acetone was serially diluted in dH20 to concentrations of 100,

33.3, 11.1, 3.7, 1.2, 0.4, 0.14, 0.04, 0.015, and 0.005 pg/ml and added to a microtiter plate contaning 1.0 ng/well OVA-Phe-CDA. RAX11, diluted 1:20,000 in PBST, was added and the plate incubated 60 min at 37 °C. After washing the plate of unbound antibodies, goat-anti rabbit IgG-HRP, diluted 1:10,000 in PBST, was added and incubated 30 min at

37 °C. The TMB enzyme substrate was incubated 30 min at room temperature and the reaction stopped with 4M H2S04. The linear range of CDA detection was 0.015 to 10 ug/ml with an IC50 value of 2 ug/ml. c. Cross-reactivity to related compounds

The cross-reactivities of RAX11 to compounds structurally related to CDA (Fig. 13) were determined by competition assays over the concentration range 100 to 0.4 pg/ml.

The IC50 values were extrapolated from plots of %B/Bo versus concentration. Percent cross reactivity (%R) was quantitated as:

%R = [ICso (CDA) / IC50 (test compound)] x 100.

54 RAX11 specifically detected CD A and showed no cross reactivity to alachlor, other alachlor metabolites, or other chloroacetanilide herbicides (Table 9). Thus, the immunogen BSA-Phe-CDA resulted in the production of CDA-specific antibodies.

Table 8. Checkerboard Assay for the Detection of OVA-Phe-CDA by Antisera RAX10 and RAX11

OVA-Phe-CDA: 0,5 ng/well_1.0 ng/well 5.0 ng/well 10.0 ng/well

2° Ab: 1:5K 1:10K 1:20K 1 1:5K 1:1 OK 1:20K 1 1:5K 1:10K 1:20K 1 1:5K 1:10K 1:20K

RAX10 1:5K 1.020 1 0.829> 0.701 1.159 1.090 0.889 1.524 1.414 1.223 1.535 1.448 1.218

1:1 OK 1.000 0.815 0.646 1.103 1.020 0.851 1.475 1.282 1.186 1.446 1.402 1.310

1:20K 0.837 0.734 0.576 1.171 0.955 0.763 1.394 1.309 1.091 1.460 1.380 1.187

No 1° Ab 0.056 0.047 0.041 0.044 0.042 0.042 0.044 0.042 0.042 0.044 0.043 0.048

RAX11 1:5K 0.969 0.776 0.612 1.203 1.070 0.733 1.504 1.354 1.212 1.540 1.426 1.292

1:10K 0.736 0.631 0.435 1.120 0.900 0.698 1.425 1.195 1.074 1.552 1.365 1.228

1:20K 0.755 0.597 0.426 0.963 0.842 0.611 1.312 1.138 0.955 1.399 1.214 1.072

No 1° Ab 0.056 0.049 0.044 0.052 0.052 0.052 0.053 0.058 0.048 0.056 0.053 0.047

1. OD determined at 450 nm.

55 Figure 14. Competitive inhibition of antiserum RAX11 by CD A. Microtiter plates were coated with 1 ng/well OVA-Phe-CDA. Antiserum RAX11 was diluted 1:20,000 in PBST.

Goat anti-rabbit IgG was diluted 1:10,000 in PBST. Absorbance values (OD) are the mean of eight replicates. Error bars represent ± 1 SD.

56 0.6

0.5 -

0.4 -

0.1 -

0.0 H— 1 1111 1 i i i i i 111 i i i i i ii 11 i i r i i i 111-1-1—i i i i 111- 0.000 0.010 0.100 1.000 10.000 100.000 CDA Concentration (ug/ml)

57 Table 9. Cross-Reactivity of RAX11 to Chloroacetanilide Herbicides and Related Compounds

%R3 Compound Concentration %B/Bo' ca. IC502

<0.1% Alachlor 100 pg/ml 86.2% > 100 jug/ml <0.1% 2,6-DEA 100 jLig/ml 100.6% >100 jug/ml <0.1% Hydroxyalachlor 100 pg/ml 98.9% > 100 jag/ml <0.1% HDA 100 pg/ml 91.0% > 100 jag/ml

ESA-alachlor 100 pg/ml 94.8% > 100 pg/ml <0.1%

Acetochlor 100 pg/ml 93.4% > 100 pg/ml <0.1%

Butachlor 100 jug/ml 99.9% > 100 jug/ml <0.1%

Metolachlor 100 pg/ml 88.3% > 100 pg/ml <0.1%

Propachlor 100 pg/ml 101.6% >100 pg/ml <0.1%

1. Percent ratio of bound antibody compared to dH20 control. 2. Concentration resulting in 50% inhibition of antibody binding versus dH20 control. 3. Cross reactivity compared to CDA. %R=[IC5o (CDA) / IC50 (test compound)] x 100.

d. Assay Optimization

i. pH effects. The effect of pH on assay performance was determined with pH adjusted dH20 over the range pH 4.0 to 9.5 in 0.5 pH increments. The standard cELSIA format was run with no analyte present (i.e., pH adjusted dH20 only) or with a concentration of

CDA that resulted in 50% inhibition of antibody binding (i.e., 2 pg/ml). The inhibition of

antiserum RAX11 by CDA was unaffected by pH over the entire range tested (Fig. 15).

58 Figure 15. Effect of aqueous sample pH on binding of antiserum RAX11.

59 Sample pH Value

60 ii. Solvent effects. To determine the effect of solvents on assay performance, the standard cELSIA format was run with standards of CD A over the concentration range 0.1 to 10 pg/ml made up in 10%, 25% or 50% methanol, ethanol, acetone or acetonitrile in dH20. Methanol and ethanol concentrations up to 10% did not adversely effected the signal generated (Table 10). However, higher concentrations of these solvents, and as little as 10% acetone or acetonitrile, resulted in a reduction in OD signal reduction of 20 to 60%. Therefore, use of these solvents for extraction processes prior to anlysis by cELISA 3 would require solvent removal and reconstitution in dH20 or buffer.

Table 10. cELISA3 Signal Attenuation Due to the Effect of Solvents on Antibody Binding

Solvent Concentration Percent Signal Reduction1

Methanol Ethanol Acetone Acetonitrile

10% (v/v) 2.6 ±2.3 11.5 + 5.8 19.8 ± 10.0 26.9 ± 10.8

25% 18.4 ±2.4 37.2 + 3.6 21.4 ± 11.8 39.8 ±9.6

50% 20.5 ±3.1 61.4 + 3.3 17.2 ±8.6 57.7 ±9.7

1. Percent reduction in OD compared to dH20 control. Values are averages (n = 8) ± standard deviation of duplicate determinations over the concentration range 0.1- 10 pg/ml.

61 4. cELISA 4: Anti-BSA-AHT-CDA Antisera a. Checkerboard assays

Microtiter plates were coated with OVA-AHT-CDA at 1.0, 10.0, 25.0 and 50.0 ng/well.

Antisera RAX8 and RAX9 were serially diluted 1:5000, 1:10,000 and 1:20,000 in PBST.

RAX8 showed a high CDA-specific antibody titer and was diluted 1:25,000, 1:50000,

1:100,000 and 1:200,000 in susequent checkerboard assays. Goat anti-rabbit IgG-HRP was serially diluted 1:10,000 and 1:20,000 in PBST. Assays were run as described in section III.B. Results for RAX8 are shown in Table 11. b. Competition assays

Initial competition assays showed that both RAX8 and RAX9 specifically recognized

CDA and HDA in solution. RAX8 produced a more satisfactory competition curve and was used in all subsequent competition assays (Fig. 16) A stock solution of 10 mg/ml

CDA in acetone was serially diluted in dH20 to concentrations of 100, 33.3, 11.1, 3.7,

1.2, 0.4, 0.14, 0.04, 0.015, and 0.005 pg/ml and added to a microtiter plate containing 1.0 ng/well OVA-AHT-CDA. RAX8, diluted 1:50,000 in PBST, was added and the plate incubated 60 min at 37 °C. After washing the plate of unbound antibodies, goat-anti rabbit IgG-HRP, diluted 1:10,000 in PBST, was added and incubated 30 min at 37 °C.

The TMB enzyme substrate was incubated 30 min at room temperature and the reaction stopped with 4M H2S04. The linear range of detection for CDA and HDA was 0.01 to

10 ug/ml with an IC50 value of 0.2 ug/ml.

62 Table 11. Checkerboard Assay for the Detection of OVA-AHT-CDA by Antiserum RAX8

1° Ab: 1:25,000_1:50,000_1:100,000 1:200,000

2° Ab 1:1 OK 1.0201 0.829 0.701 1.159 1.090 0.889 1.524 1.414 1.223 1.535 1.448 1.218

1.000 0.815 0.646 1.103 1.020 0.851 1.475 1.282 1.186 1.446 1.402 1.310

0.837 0.734 0.576 1.171 0.955 0.763 1.394 1.309 1.091 1.460 1.380 1.187

No 1° Ab 0.056 0.047 0.041 0.044 0.042 0.042 0.044 0.042 0.042 0.044 0.043 0.048

1:20K 0.969 0.776 0.612 1.203 1.070 0.733 1.504 1.354 1.212 1.540 1.426 1.292

0.736 0.631 0.435 1.120 0.900 0.698 1.425 1.195 1.074 1.552 1.365 1.228

0.755 0.597 0.426 0.963 0.842 0.611 1.312 1.138 0.955 1.399 1.214 1.072

No 1° Ab 0.056 0.049 0.044 0.052 0.052 0.052 0.053 0.058 0.048 0.056 0.053 0.047

1. Optical density at 450 nm.

c. Cross-reactivity to related compounds

The cross-reactivity of antiserum RAX8 to compounds structurally related to CDA and

HDAwas determined as described in section III.C.3.iii. RAX8 showed minimal cross¬ reactivity to alachlor, hydroxyalachlor and 2,6-diethylaniline, but did not cross-react with the chloroacetanilide herbicides acetochlor, butachlor, metolachlor and propachlor (Table

12). Thus, the immunogen BSA-AHT-CDA resulted in the production of antibodies specific for CDA and HDA.

63 Figure 16. Competitive inhibition of antiserum RAX8 by CD A and HD A. Microtiter plates were coated with 1 ng/well OVA-AHT-CDA. Antiserum RAX8 was diluted

1:50,000 in PBST. Goat anti-rabbit IgG was diluted 1:10,000 in PBST. Absorbance values (OD) are the mean of eight replicates. Error bars represent ± 1 SD.

64 1.50

1.25

1.00 -

0.75

0.50 -

0.25

0.00 I I I I I I I I-1—I TTTTT 1 I I n I I I I I I-1—I TTTTT]-1-1—r I I I I 0.000 0.010 0.100 1.000 10.000 100.000 Concentration (ug/ml)

65 Table 12. Cross-Reactivity of Antiserum RAX8 to Chloroacetanilide Herbicides and Related Compounds

Compound IC5o' %R2

Alachlor 28 jig/ml 0.7%

2,6-DEA 10 jug/ml 2%

Hydroxyalachlor 28 pg/ml 0.7%

ESA-alachlor > 100 jug/ml <0.2%

Acetochlor > 100 pg/ml <0.2%

Butachlor >100 jag/ml <0.2%

Metolachlor > 100 jug/ml <0.2%

Propachlor >100 jug/ml <0.2%

1. Concentration resulting in 50% inhibition of antibody binding versus dH20 control. 2. Cross reactivity compared to CDA. %R=[IC50 (CDA) / IC50 (test compound)] x 100.

d. Assay optimization

CDA that resulted in 50% inhibition of antibody binding (i.e., 0.2 jag/ml). The inhibition of antiserum RAX8 by CDA was optimal between pH 7-8, but was not adversely affected by pH over the entire range tested (Fig. 17).

iL Solvent effects. The effect of solvents commonly used in sample extraction on assay performance was determined by diluting CDA standards in the concentration range 0.1 -

10 pg/ml in 10 to 50% solutions of methanol, ethanol, acetone.

66 Figure 17. Effect of aqueous sample pH on binding of antiserum RAX8.

67 Sample pH

68 or acetonitrile / CIH2O. Assays were performed using the standard format. Methanol and

acetone did not adversely effect assay performance, resulting in only 5 - 15% signal loss

at concentrations up to 50% (Table 13). Ethanol resulted in a 25 - 64% loss of signal,

most likely due to denaturing of antibody protein. Acetonitrile resulted in a 31 - 38%

signal loss. In summary, up to 50% sample concentrations of methanol or acetone are

acceptable, however ethanol or acetonitrile should be avoided for extraction purposes or

solvent-switched before the assay.

Table 13. cELISA4 Signal Attenuation Due to the Effect of Solvents on Antibody Binding

Solvent Concentration Percent Signal Reduction1

Methanol Ethanol Acetone Acetonitrile

10% (v/v) 7.8 25.2 14.1 38.2

25% 4.7 30.3 14.7 31.1

50% 5.7 63.4 12.9 37.6

1. Percent reduction in OD compared to dH20 control. Values are averages of duplicate determinations with coefficients of variation less than 10%.

69 D. cELISA Validation

1. Solid Phase Extraction of Aqueous Samples

CDA and HDA were extracted from aqueous samples using Oasis™ HLB (3cc, 60 mg)

solid phase extraction cartridges (Waters, Medford, MA). Fortified samples were

prepared by fortifying tap water with stock solutions of CDA or HDA (1.0 mg/ml in

acetone). The cartridges were conditioned with 6 ml of methanol followed by 6 ml dH20.

Aqueous samples (100 - 1000 ml) were loaded onto the columns at a flow rate of 5 - 10

ml/min. The cartridges were dried by maintaining the vacuum for 5 min and CDA or

HDA residues were eluted with 1.0 ml methanol. Both compounds were quantitatively

recovered from fortified aqueous samples in the range 10- 500 ng/L (Table 14). The

extracts were diluted 1:4 in dH20 for cELISA analysis or injected undiluted for analysis

by GC/MS.

Solid phase extraction of 1 L aqueous samples results in an analyte concentration factor

of 1000. When this sample preconcentration is used, the limit of detection for CDA by

CELISA3 is 15 ng/ml / 1000 = 15 pg/ml or 15 parts per trillion. The limit of detection for

CDA and HDA by cELISA4 is 10 pg/ml. The correlation of fortified aqueous sample

concentrations of CDA and HDA to concentration determined by cELISA3 and cELISA4

in the range 0.01 to 1 ng/ml showed good agreement (Figs. 18 and 19). For cELISA3, the slope of the CDA correlation line was 0.951 with an r2 value of 0.991. For cELISA4 , the slope of the CDA line was 0.996 with an r2 value of 0.996. The slope of the HDA line was 0.764 with an r2 value of 0.991. Therefore, the combination of solid phase

70 extraction and cELISA detection was suitable for the part per trillion analysis of CDA and

HD A from water samples.

Table 14. Recovery of CDA and HD A from Fortified Water Samples by Solid Phase Extraction

Fortification Level (ng/L) CDA % Recovery1 HDA % Recovery

10.0 107.3 ± 16.0 88.0 ±4.5

20.0 100.0 ±25.8 85.1 ±7.0

50.0 113.2 ±1.7 88.5 ± 1.7

500.0 113.8 ± 11.0 87.0 ±4.8

1. Mean value ± standard deviation, n=3.

71 Figure 18. Correlation of fortified sample CDA concentration with CDA concentration detected by cELISA3.

72 0.4 0.6 0.8 1.0 Fortified Concentration (ng/ml)

73 Figure 19. Correlation of fortified sample CD A and HD A concentration with CD A and

HD A concentration detected by cELISA4.

74 Concentration (ng/ml) Determined by SPE / cELISA 75 2. Chromatographic Analysis of Aqueous Sample Extracts

Extracts of aqueous samples were analyzed by gas chromatography on a Hewlett-

Packard HP5890 Series II equipped with an HP5971 Mass Selective Detector. The column was a 30 M x 0.25 mm DB-5, 0.25 ju film (J&W, Folsom, CA) programmed

from 100 °C (1 min hold) to 250 °C (0.5 min hold) at 20 °C/min. The injector temperature was 250 °C and the transfer line was held at 280 °C. Injection volume was 2 pi. The

detector was operated in the selected ion monitoring mode (SIM) for ions 225, 176 and

147 (CDA) and 207, 176 and 147 (HDA).

3. Correlation of cELISA and Chromatographic Analysis of Aqueous Samples

CDA and HDA fortified aqueous samples were split for analysis by GC/MS and by

cELISA3 (CDA-specific) and cELISA4 (CDA/HDA specific) to determine the accuracy

of the cELISAs. Samples were prepared by solid phase extraction (section III.D.l) for

chromatographic analysis. Samples in the concentration range 0.1 to 10 pg/ml were

analyzed directly by cELISA. Samples in the concentration range 0.01 to 10 ng/ml were

prepared by solid phase extraction as above. cELISA3 showed close correlation with

GC/MS results for the detection of CDA (slope = 0.933 ; r2 =0.992; Fig. 20). cELISA4

also showed a close correlation with GC/MS results for CDA (slope = 0.834; r2 =0.998;

Fig. 21), but it underestimated HDA concentration (slope = 0.579; r2 =0.990; Fig. 21).

4. cELSIA Analysis of Alachlor-Contaminated Groundwater Samples for CDA

Groundwater previously determined to be contaminated with alachlor was analyzed by

cELISA3 for the presence of CDA. Two samples were from a routine groundwater

76 monitoring project in Massachusetts and one sample was a composite of archived water samples from the USGS Laboratory in Lawrence, Kansas. Sample volumes were limited, therefore it was not possible to perform the CDA/HDA analysis by cELISA4. The samples were prepared by solid phase extraction as described above. CDA was detected in the two Massachusetts samples but not in the Kansas sample (Table 15). Results were confirmed by GC/MS analysis. The limited size of the Kansas sample raised the detection limit by a factor of 10, limiting the ability to determine the presence of CDA in this sample.

This limited analysis of real-world samples demonstrates the ability of cELISA3 to specifically detect CDA in groundwater samples.

Table 15. Analysis of Alachlor-Contaminated Groundwater Samples

Sample Sample Volume CDA Concentration (ng/ml)1

CELISA3 GC/MS

Mass. C7308 500 ml 0.024 ± 0.007 0.020 ± 0.004

Mass. B7557 500 ml 0.015 + 0.009 0.016 + 0.005

Kansas USGS 75 ml n/d2 n/d

1. Average concentration + SD, n=2. 2. None detected at detection limits of 0.15 ng/ml (cELISA3) and 0.10 ng/ml (GC/MS).

77 Figure 20. Correlation of CD A detection by cELISA3 with GC/MS.

78 CDA Concentration (ug/ml) Determined by GC/MS

79 Figure 21. Correlation of CD A and HD A detection by cELISA4 with GC/MS.

80 Concentration (ug/ml) Determined by GC/MS

81 CHAPTER IV

DISCUSSION AND CONCLUSIONS

A. Hapten Design

The target analytes CDA and HDA are small molecules (MW 225 and 207,

respectively) and are not immunogenic. It was necessary, therefore, to attach them to

large carrier protein molecules in order to elicit the immunogenic response that would

result in the production of specific antibodies. Antibody specificity encompasses not only

recognition of the intended target analyte but also a lack of cross-reactivity to structurally related compounds (i.e., the parent compound alachlor, other alachlor metabolites, and other chloroacetanilide herbicides). The attachment of CDA or HDA to protein molecules necessitated the synthesis of chemical derivatives (i.e., haptens) that would provide a covalent linkage as well as present the hapten for antibody recognition in a manner that maximized the unique structural features of CDA and HDA.

The structural differences between CDA and HDA, alachlor, other alachlor metabolites, and other chloroacetanilide herbicides are: 1) the degree of saturation at the nitrogen atom and 2) alkyl substitution at the 2 and 6 positions on the aromatic ring

(Figs. 1 and 13). The parent herbicides are fully saturated at the nitrogen atom (i.e., they are tertiary amines), whereas CDA, HDA and most other metabolites are secondary amine compounds. Alachlor and butachlor (and their metabolites) have ethyl substitutions at the

2 and 6 positions of the aromatic ring. Acetochlor and metolachlor (and their metabolites) have one ethyl and one methyl substitution at these positions. Propachlor

(and its metabolites) contains a nonsubstituted aromatic ring. Therefore, the degree of

82 saturation at the nitrogen atom and the type of alkyl substitution on the aromatic ring would be expected to govern any potential cross-reactivity to CDA and HDA specific antibodies.

HDA can be synthesized from CDA by a simple hydrolysis reaction. Therefore, hapten synthesis focused on CDA, with the recognition that the corresponding HDA hapten could be synthesized in a similar manner with the addition of a hydrolysis step to the reaction sequence. There were four candidate sites on CDA for use as covalent binding sites to the carrier protein: 1) the nitrogen atom, 2) the chloroacetanilide chlorine atom, 3) the ethyl side chains of the aromatic ring and 4) the aromatic ring itself (Fig. 22).

The C2-CDA and C4-CDA haptens featured a 2- or 4-carbon linkage group, respectively, attached to the nitrogen atom. The use of the nitrogen atom of CDA as the linkage site had the advantage of straight-forward synthetic procedures to produce the desired hapten. Anitsera produced in response to the BSA-C2-CDA immunogen demonstrated greater affinity for alachlor than for CDA. The length of the linkage group also influences hapten presentation for antibody recognition (Goodrow et al. 1995).

Therefore, the BSA-C4-CDA was synthesized with a 4-carbon spacer to replace the 2- carbon spacer of C2-CDA. Anitsera produced in response to BSA-C4-CDA showed a greater affinity to CDA than did BSA-C2-CDA antibodies. However, it still elicited far greater affinity for alachlor. This finding demonstrated clearly that the saturation state of the amine moiety was crucial for selective antibody recognition. By utilizing the nitrogen atom of CDA for the linkage site, the resulting hapten contained a tertiary amine moiety that more closely resembled alachlor than the secondary amine CDA. It is now apparent

83 Figure 22. Candidate attachment sites for covalent linkage of CDA to carrier protein molecules.

84 85 that this hapten synthesis strategy is unsuitable for the production of CDA or HDA

specific antibodies.

The Phe-CDA hapten featured a 5-carbon linkage group attached to carbon 4 of the

aromatic ring of CDA. This linkage strategy was used successfully by Schlaeppi et al.

(1991) for the synthesis of a haptenic derivative of metolachlor. The attachment of a

linker group directly to the aromatic ring involved a more complicated synthesis, but the resulting hapten had the advantage of preserving the saturation state of the secondary amine moiety of CDA. The length of the linkage also insured that the hapten was physically distant enough from the carrier protein surface that the alkyl side chains on the aromatic ring could also contribute to antibody recognition. Anitsera produced in response to BSA-Phe-CDA specifically recognized CDA, but did not recognize HDA.

Apparently, this antisera (i.e., RAX11) elicited adequate specificity to distinguish between the chlorine atom of CDA and the hydroxyl moiety of HDA, the sole difference between these two molecules. RAX11 also did not cross-react with alachlor, other alachlor metabolites or other chloroacetanilide herbicides. The lack of cross-reactivity to alachlor-ESA is particularly significant because this compound was found to be a major cross-reactive interference with commercial alachlor immunoassays that resulted in a number of false-positive reports for alachlor (Baker, et al. 1993). Thus, by preserving the unique acetanilide moiety of CDA in the hapten design, the goal of producing CDA specific antibodies was achieved.

The use of the chlorine atom of CDA as a linkage site avoided the synthesis of a unique hapten altogether by allowing direct coupling of CDA to BSA through the coupling reagent AHT. Feng et al. (1992) successfully utilized this approach to create hapten

86 carrier conjugates for a variety of chloroacetanilide compounds. Although the BSA-

AHT-CDA immunogen preserved the secondary amine state in the hapten, this linkage

strategy eliminated the only distinguishing structural feature between CDA and HDA (

i.e., the chlorine atom and the hydroxyl group, respectively). Therefore, antisera

produced in response to the BSA-AHT-CDA immunogen (i.e., RAX8) recognized both

CDA and HDA but distinguished these compounds from alachlor, other alachlor

metabolites and other chloroacetanilides. The minimal cross reactivity of RAX8 to

alachlor, hydroxyalachlor and 2,6-diethylaniline indicates that in addition to the saturation

state of the amine, the ethyl substitution at the 2 and 6 positions of the aromatic ring also

plays an important role in antibody recognition.

Various haptenic derivatives of an acetanilide compound, CDA, produced antisera with

differing specificities. These results clearly indicate that for the production of antisera

specific for acetanilides, the degree of saturation of the amine moiety is of primary

importance and must be preserved in the hapten design. Alkyl substitution in the aromatic portion of the molecule is of secondary importance, but does play a role in antibody recognition.

B. The CDA and CD A/HD A cELISAs

cELISA3 was based on antiserum RAX11, which specifically detected CDA. The

range of detection was 0.015 to 10 pg/ml, with an IC50 of 2 jug/ml. The detection range of cELSIA in general can be in the low part per billion range, so cELISA3 is not at the limits of immunoassay sensitivity. Additionally, a heterologous assay system was used in which different protein carriers were used for raising antisera and for the plate-coating

87 antigen. This insured that only CDA-specific antibodies would react in the assay and insured that antibodies that recognized the BSA carrier protein would not cause interference and compromise sensitivity. One possible means of enhancing the sensitivity of cELISA 3 would be through the purification of antiserum RAX11 by affinity chromatography (Szekacs and Hammock 1995).

Solid phase extraction procedures were developed for extraction of CDA from aqueous samples. It was possible to effect a 1000-fold concentration of CDA. Thus, the effective detection limit for CDA in aqueous samples was lowered to 15 parts per trillion (pg/ml) by SPE/cETISA3 analysis. Also, cELISA3 did not cross-react with HD A, alachlor or its other metabolites, or other chloroacetanilide herbicides.

Groundwater or surface water samples exhibit a range of pH values, so it was important to assess the influence of pH on the assay. The pKas of CDA and HDA are unknown, but for the parent aniline the pKa > 27 (Harris and Hayes 1982). Therefore, the amine moieties of CDA and HDA most likely are ionized only in very acidic conditions, so any potential influence of pH on the assay would be on antibody binding rather than on the ionization state of the target analytes. cELISA3 was shown to be stable to changes in sample pH, due to the buffering effects of the antiserum solution buffer.

Polar solvents are utilized for the extraction of CDA from soil and other environmental samples, as well as in the solid phase extraction procedures used for aqueous samples. cELSIA3 is adversely effected by the solvents acetone and acetonitrile, but is stable to low concentrations (i.e., 10% or less) of methanol and ethanol.

cELISA4 was based on antiserum RAX8, which specifically detected CDA and HDA.

The range of detection was 0.01 to 10 (ag/ml, with IC50s of 0.5 (.ig/ml for CDA and 0.3

88 pg/ml for HDA. This assay also used a heterologous design, with different carrier proteins for the immunizing and plate-coating conjugates. Identical competition curves for CDA and HDA would be expected based on the hapten design. The slight differences in the shapes of the competition curves and for the IC50S are likely due to differences in the water solubilities of the two compounds. Other contributing factors are steric differences between CDA and HDA (i.e., the chlorine of CDA versus the hydroxyl group of HDA), and the response of different portions of the polyclonal antibody pool to each compound.

The solid phase extraction procedures developed for CDA were also effective for HDA.

The effective detection limit for the combined SPE/cELISA4 analysis of CDA and HDA from aqueous samples was 10 parts per trillion (pg/ml). cELISA4 did not cross-react with alachlor, its metabolites or other chloroacetanilide herbicides. It was stable to changes in sample pH, due to the buffering effects of the antiserum solution buffer. It was adversely effected by the solvents ethanol and acetonitrile, but was stable to sample concentrations of up to 50% methanol or acetone.

C. Conclusions

Two cELISAs have been developed for the specific detection of 2-chloro-2’,6’-diethylacetanilide (CDA) and 2-hydroxy-2’,6’-diethylacetanilide (HDA), two mutagenic metabolites of the chloroacetanilide herbicide alachlor. cELISA3 is specific for CDA with a detection range of 0.015 to 10 pg/ml. Solid phase extraction of CDA residues from aqueous samples gives a 1000-fold concentration factor resulting in an effective detection limit of 15 pg/ml. cELISA4 is specific for both CDA and HDA in combination, with a

89 detection range of 0.01 to 10 jug/ml. Solid phase extraction of aqueous samples prior to cELISA analysis results in an effective detection limit of 10 pg/ml. Chloroacetanilide herbicides and other alachlor metabolites that may be present in environmental samples do not interfere with the detection of CD A and HD A.

The CDA and CDA/HDA cELISAs provide the means of monitoring for these two mutagens in the environment. The availability of the CDA and CDA/HDA specific antisera provides a means for analyte-specific preconcentration of samples for use in traditional chromatographic analytical techniques.

90 APPENDICES

91 APPENDIX A

MASS SPECTRA < Cj u s W) S

93 94 95 96 97 98 fO o ccS o ’■4—* O

-4—<

-5i <41 o

c/2 -*-» C 3 c/2 C/2 CCj

ir>o un O s 2 -4—>o

99 100 101 h o r~ o ca - co CM

CA - o r- CO CM CM CO CD CM h O CD CM CO

CM Q - CM

O CA CD CM CM CM < CM J. CM < O 4 0 CM r° r cm L r-1 ~CQ- o -H 1-1 CO E =4- ca LD o _j- o CD CD 1—I 1—I CD

CM CD CO I> C (IS o CO

CA r CM o

CD 0 0 O O O O c 0 O O O O cC 0 O O O O A TS 0 O O O O I a 0 O O O O - Z / H 3 LD CO CM i—1

102 103 Abundance Scan 764 (11.459 min): ALA001. 104 o c

3i mr\ CmO

c/2 *5 3 C/2 C/2 Ctf s o tn l/"2 oi

B S -4—>o

r- 4/ U 3 WD £

105 -H CO ro LD m co i—i Q S CO Eh en e'¬ p U 03 C (Abundance £ fD ID co o o o \—1 o 1 '’ ^ '1 CO o o o o rH CM \—I \—I ' ’I o o o o rH in o T—I -T—T—p-r—r CM \—1 o o o o -T—r-p-r-r- o o o \—1 o o 106 ~t—«—j—i—r CO o o o o r“' 1 CO o o o o G\ <—I -T—r-p-r-r- i> O o o o o ■' I1 CM 1—I CM d o r- t-1 o CO in CM o o o o in \—I t-1-i-r— -CO— CO | co cn O

m/z--> _50 60 70 80 90 100 110 120 130 140 150 107 Abundance Scan 527 (9.602 mm) : DMT726.D 25000 -I 1$0 o CM o o o T O O LD O \—1 O O O O t—1 o o o o N A f i 109 110 VO r i o

£ 2 •4o—» o a, C/2 C/2 ct3C/2 £ c OC3 C/2 3

111 112 'O rs ^

QJ 0) 6 ^ o c -»—>oJ o

< a> u 3 WD S

113 in ld — co

CM CO

l—1 p • F* CM CM O o E-f S Q \—1 og a 00 -H \-1 E PM O CM cr> CM CO

CM t> £ o C/3 ln T-f i—i

LO LD

-r -1-r-1-.-i- 1 '' 1 1 ^ (D r_ —1— ' ^ 1 r U o o o o o a o o o o o 03 o o o o o T5 o o o o o £ o o o o o 3 If) CO CM T—1

a m/z- - >_50_ 100 150 _ 200 250 300 350 APPENDIX B

INFRARED SPECTRA

115 ■t—»

% -C ■+->o ~3I VO (N O

B 2 ■4—>o

116 3500 ' ~ ‘ 3000 (CiV: ') 2500 2000 1800 1600 1400 1200 1000 800 (CM 600 400

117 c4 CQ 4> u 3 D£ £

118 119

O a o

-C ■*—>

CQ o '» 3 s&£

120 121 122 123 124 125 126 1_L 127 128 129 130 131 132 JL-M- 133 < Q U

C/3

O od

£ 2 -4—> o

134 135 < Q U I o 4-1 o Uh • H O d

PQ Cl u 3 W) E

136 loOCflCM ’) 2500 2000 1800 1600 MOO 1200 1000 (CM 1 ) 800 600 400

137 138 139 APPENDIX C

‘H NUCLEAR MAGNETIC RESONANCE SPECTRA

140 141 PPM

142 143 o o

\ o > i- • PPM i J

If t r 1i

j- o CD

144 BIBLIOGRAPHY

Aizawa, H., 1982. Metabolic Mans of Pesticides. New York, Academic Press.

Baker, D. B., R. J. Bushway, S. A. Adams , C. Macomber, 1993. Immunoassay screens for alachlor in rural wells: false positives and an alachlor soil metabolite. Environ. Sci. Tech. 27(3): 562-564.

Barbash, J. E., E. A. Resek, 1996. Pesticides in Groundwater: Distribution. Trends, and Governing Factors. Chelsea, MI, Ann Arbor Press.

Barrett, M. R., 1996. The environmental impact of pesticide degradates in groundwater. in Hebicide Metabolites in Surface Water and Groundwater. M. T. Meyer and E. M. Thurman, eds. ACS Symposium Series No. 630. Washington, DC, American Chemical Society.

Brady, J. F., 1995. Interpretation of immunoassay data, in Immunoanalysis of Agrochemicals. J. O. Nelson, A. E. Karu, R. B. Wong, eds. ACS Symposium Series No. 586. Washington, DC, American Chemical Society.

Chesters, G., G. V. Simsiman, J. Levy, B. J. Alhajar, R. N. Fathull, J. M. Harkin, 1989. Environmental fate of alachlor and metolachlor. in Rev. Environ. Contam. & Toxicol. v. 110. New York, Springer-Verlag.

Coligan, J. E., A. D. Kruisbeek, D. H. Margulies, E. M. Shevach , W. Strober, 1995. Current Protocols in Immunology. New York, Wiley Interscience.

EPA, 1987. Alachlor Health Advisory. Washington, DC, U.S. Environmental Protection Agency Office of Drinking Water.

Feng, P. C. C., S. R. Horton , C. R. Sharp, 1992. A general method for developing immunoassays to chloroacetanilide herbicides. J. Agric.Food Chem. 40(2): 211-214.

Gianessi, L. P., J. E. Anderson, 1995. Pesticide Use in U.S. Crop Production: National Summary Report. Washington, DC, National Center for Food and Agricultural Policy.

Goodrow, M. H., J. R. Sanborn, D. W. Stoutamire, S. J. Gee, B. D. Hammock, 1995. Strategies for immunoassay hapten design, in Immunoanalysis of Agrochemicals. J. O. Nelson, A. E. Karu, R. B. Wong, eds. ACS Symposium Series No. 586. Washington, DC, American Chemical Society.

Goolsby, D. A., E. M. Thurman, M. L. Pomes, M. T. Meyer , W. A. Battaglin, 1997. Herbicides and their metabolites in rainfall: origin, transport and the deposition patterns across the midwestem and northeastern United States, 1990-1991. Env. Sci. Tech. 31(5): 1325-1333.

145 Habeeb, A. F. S. A., 1966. Determination of free amino groups in proteins by trinitrobenzenesulfonic acid. Anal. Biochem. 14:328-336.

Harlow, E., D. Lane, 1988. Antibodies: A Laboratory Manual. Cold Spring Harbor, NY, Cold Spring Harbor Laboratory.

Harris, J.C., M.J. Hayes, 1982. Acid Dissociation Constant, in Handbook of Chemical Property Estimation Methods. W.J. Lyman, W.F. Reehl, D.H. Rosenblatt, eds. New York, McGraw Hill.

Hayashi, M., T. Morita, Y. Kodoma, T. Sofuni, M. Ishidate, 1990. The micronucleus assay with mouse peripheral blood reticulocytes using acridine orange coated slides. Mut. Res. 245: 245-249.

Henion, M. C., V. Pichon, M. Bouzige, 1997. Immunochemistry for the selective pretreatment of environmental samples. Paper no. 43. 214th National Meeting of the American Chemical Society, Las Vegas, NV.

Heyer, R., H. J. Stan, 1995. Comparison of the leaching behaviour of alachlor and its metabolites under field and laboratory conditions. Intern. J. Environ. Anal. Chem. 58: 177-183.

Hoberg G., 1990. Risk, science and politics: alachlor regulation in Canada and the United States. Can. J. Pol. Sci. 23(2):257-277.

Jacoby, H., C. Hoheisel, J. Karrie, S. Lees, L. Davies-Hilliard, P. Hannon, R. Bingham, E. Behl, D. Wells , E. Waldman, 1992. Pesticides in Ground Water Database. Washington, DC, EPA.

Jakoby, W. B. 1978. The glutathione S-transferases: a group of multifunctional detoxification enzymes, in Advances in Enzymology. A. Meister, ed. New York, J Wiley.

Lee, R. F., J. B. Weber, 1993. Influence of polymers on the mobility, loss and bioactivity of 14C from 14C-labeled atrazine, metolachlor and primisulfuron. J. Agric. Food Chem. 41(6): 988-995.

Maron, D. M., B. N. Ames, 1983. Revised methods for the Salmonella mutagenicity test. Mut. Res 113: 173-215.

Montgomery, J. H., 1993. Agrochemicals Desk Reference. Ann Arbor, Lewis Publishers.

Novick, N. J., R. Mukherjee , M. Alexander, 1986. Metabolism of alachlor and propachlor in suspensions of pretreated soils and in samples from groundwater aquifers. J. Agric. Food Chem. 34(4): 721-725.

146 Pereira, W. E., F. D. Hostettler, 1993. Nonpoint source contamination of the Mississippi River and its tributaries by herbicides. Environ. Sci. Tech. 27(8): 1542-52.

Pereira, W. E., C. E. Rostad , T. J. Leiker, 1992. Synthetic organic chemicals in the lower Mississippi River and its major tributaries: distribution, transport and fate. J. Contain. Hydrol. 9: 175-188.

Pereira, W. E., C. E. Rostad, 1990. Occurrence, distribution and transport of herbicides and their degradation products in the lower Mississippi River and its tributaries. Environ. Sci. Tech. 24(9): 1400-1406.

Pereira, W. E., C. E. Rostad , T. J. Leiker, 1990. Determination of trace levels of herbicides and their degradation products in surface and ground waters by gas chromatography / ion trap mass spectrometry. Anal. Chim. Acta. 228: 69-75.

Potter, T. M., T. M. Carpenter, 1995. Occurrence of alachlor degradation products in groundwater. Environ. Sci. Tech. 29(6): 1557-1563.

Schlaeppi, J. M., H. Moser , K. Ramsteiner, 1991. Determination of metolachlor by competitive enzyme immunoassay using a specific monoclonal antibody. J. Agric. FoodChem. 39(8): 1533-1536.

Sojka, R. E., R. D. Lentz, 1993. Erosion takes a powder. Agric. Res. 41(9): 16-17.

Sun, M., 1986. EPA proposal on alachlor nears. Science. 233:1143-1144.

Szekacs, A., B. D. Hammock, 1995. Development of an enzyme-linked immunosorbent assay for the detection of the triazole fungicide mycobutanil. J. Agric. Food Chem. 43(8): 2083-2091.

Teidje, J. M., M. L. Hagedom, 1975. Degradation of alachlor by a soil fungus, Chaetomium globosum. J. Agric.Food Chem. 23(1): 88-81.

Tessier, D. M., J. M. Clark, 1995. Quantitative assessment of the mutagenic potential of environmental degradative products of alachlor. J. Agric. Food Chem. 43(9): 2504-12.

Tessier, D. M. 1994. Environmental Degradative Products of the Herbicide Alachlor: Quantitative Assessment of Mutagenic Potential. Master’s Thesis. University of Massachusetts Amherst.

Thurman, E. M., D. A. Goolsby, D. S. Aga, M. L. Pomes , M. T. Meyer, 1996. Occurrence of alachlor and its sulfonated metabolite in rivers and reservoirs of the midwestem United States; the importance of sulfonation in the transport of chloroacetanilide herbicides. Environ. Sci. Tech. 30(2): 569-574.

147 Tijssen, P„ 1985. Practice and Theory of Enzyme Immunoassays. New York, Elsevier.

Weitersen, R. C., T. C. Daniel, K. J. Fermanich, B. Lowery , K. McSweeney, 1993. Irrigation and polymer effects on herbicide transport through the unsaturated zone of a Sparta sand. J. Env. Qual. 22: 819-824.

Wienhold, B. J., T. J. Gish, 1994. Effect of Formulation and Tillage Practice on Volatilization of Atrazine and Alachlor. J. Env. Qual. 23(2): 292-298.

148